The evidence for your lack of technical knowledge when you facked-up basic PS4's memory bandwidth math.
You misuse quotes i.e. memory modules operates at certain bits and clock speed. You didn't factor in crappy memory controllers from AMD.
If PS4's GPU delivers 172 GB/s while PC GPUs has ~79 percent efficiency, PS4's 32 ROPS vs memory bandwidth would be rivaling 7950-BE's 32 ROPS vs memory bandwidth. This is NOT the case. 7950-BE has a better MSAA 4X performance (32 ROPS and memory bandwidth) than PS4.
PS4 is inferior to 7950, 7950-BE/8950 OEM, 7870 XT, 7870 GE, R9-270X, R9-270, R7-370 and higher GPU SKUs.
The evidence for your lack of technical knowledge when you facked-up basic PS4's memory bandwidth math.
You misuse quotes i.e. memory modules operates at certain bits and clock speed. You didn't factor in crappy memory controllers from AMD.
If PS4's GPU delivers 172 GB/s while PC GPUs has ~79 percent efficiency, PS4's 32 ROPS vs memory bandwidth would be rivaling 7950-BE's 32 ROPS vs memory bandwidth. This is NOT the case. 7950-BE has a better MSAA 4X performance (32 ROPS and memory bandwidth) than PS4.
PS4 is inferior to 7950, 7950-BE/8950 OEM, 7870 XT, 7870 GE, R9-270X, R9-270, R7-370 and higher GPU SKUs.
Evidence of your lack of knowledge...Hahahaaaaaaaaaaaaaaaaaaaaaaaa
The xbox one can't use 68GB/s + 104GB/s because CPU and system take 30GB/s from the main pool leaving only 38GB/s...lol
That fossy math of yours...
PC is PC the PS4 is a console memory controllers are different,in fact does GPU eat CPU bandwidth desproportionally on PC.?
No wait that only happen on PS4 because it uses GDDR5 for memory and has a CPU on the same die which feeds from it,so the memory controller is different.
The 7950 BE has better MSAA than the PS4 because it has freaking 28 CU,384 bit bus and higher bandwidth than the PS4 you freaking fool,What the fu** man do you really believe that because a damn GPU has more bandwidth than another that mean it most perform better or equal without taking into notice power.?
This is another one that is getting quote..
PS4 176GB/s shared, 18 CU 256 Bit bus.
PC 7950 BE 240GB/s solely for GPU, 28 CU and 384 bit bus.
Explain to me how even if the PS4 was able to use all 176GB/s for its GPU (which can't) how the fu** it would run MSAA 4X better or even close to a 7950 BE you blind biased fool,when the 7950 BE has 240GB/s for the GPU alone,without counting its wider 384 bit bus and extra 10 CU.?
How the fu**..
You say the XBO has bandwidth similar to Pitcairn lets test that theory....
It's clear that that both versions of UFC still operate using 1600x900 framebuffers in combination with multi-sampling anti-aliasing. As we saw in the demo code, 4x MSAA is present on PS4, reduced to 2x MSAA on the Xbox One,
The evidence for your lack of technical knowledge when you facked-up basic PS4's memory bandwidth math.
You misuse quotes i.e. memory modules operates at certain bits and clock speed. You didn't factor in crappy memory controllers from AMD.
If PS4's GPU delivers 172 GB/s while PC GPUs has ~79 percent efficiency, PS4's 32 ROPS vs memory bandwidth would be rivaling 7950-BE's 32 ROPS vs memory bandwidth. This is NOT the case. 7950-BE has a better MSAA 4X performance (32 ROPS and memory bandwidth) than PS4.
PS4 is inferior to 7950, 7950-BE/8950 OEM, 7870 XT, 7870 GE, R9-270X, R9-270, R7-370 and higher GPU SKUs.
Evidence of your lack of knowledge...Hahahaaaaaaaaaaaaaaaaaaaaaaaa
The xbox one can't use 68GB/s + 104GB/s because CPU and system take 30GB/s from the main pool leaving only 38GB/s...lol
That fossy math of yours...
PC is PC the PS4 is a console memory controllers are different,in fact does GPU eat CPU bandwidth desproportionally on PC.?
No wait that only happen on PS4 because it uses GDDR5 for memory and has a CPU on the same die which feeds from it,so the memory controller is different.
The 7950 BE has better MSAA than the PS4 because it has freaking 28 CU,384 bit bus and higher bandwidth than the PS4 you freaking fool,What the fu** man do you really believe that because a damn GPU has more bandwidth than another that mean it most perform better or equal without taking into notice power.?
This is another one that is getting quote..
PS4 176GB/s shared, 18 CU 256 Bit bus.
PC 7950 BE 240GB/s solely for GPU, 28 CU and 384 bit bus.
Explain to me how even if the PS4 was able to use all 176GB/s for its GPU (which can't) how the fu** it would run MSAA 4X better or even close to a 7950 BE you blind biased fool,when the 7950 BE has 240GB/s for the GPU alone,without counting its wider 384 bit bus and extra 10 CU.?
How the fu**..
You say the XBO has bandwidth similar to Pitcairn lets test that theory....
It's clear that that both versions of UFC still operate using 1600x900 framebuffers in combination with multi-sampling anti-aliasing. As we saw in the demo code, 4x MSAA is present on PS4, reduced to 2x MSAA on the Xbox One,
Froza H2 has MSAA 4X with forward plus rendering at 1920x1080p.
Did you forget there's a link between CPU and GPU?
Have you realized CPU's memory consumption floats between 0 to 30 GB/s?
Have you realized the purpose why you have a CPU L2 cache?
Have you realize Pitcairn ES's GDDR5 doesn't require any special programming tricks?
The 7950 BE has better MSAA than the PS4 because it has freaking 28 CU,384 bit bus and higher bandwidth than the PS4 you freaking fool,What the fu** man do you really believe that because a damn GPU has more bandwidth than another that mean it most perform better or equal without taking into notice power.?
GCN's CU has nothing to do with MSAA resolve i.e. it's ROPS unit's MSAA co-processors and memory bandwidth you stupid cow.
Using Radeon HD 4870 as an example which is basically DX10 version of Radeon HD 5770/6770 hardware.
For UFC's 1600x900p on XBO/PS4 from http://www.eurogamer.net/articles/digitalfoundry-2014-ea-sports-ufc-face-off
PS4's MSAA 4X vs XBO's MSAA 2X
however, in our demo analysis we found the PS4 code fell behind Xbox One in terms of raw stability, and sadly the situation hasn't changed at all in the final game, despite the availability of a launch-day update
Across different arenas using various combatants, we still see some minor frame-rate drops on PS4 and a few regularly torn frames (limited to the top 10 per cent of the screen, so mostly unnoticeable). The situation has little impact on how the game plays, but the dropped frames do lead to some short but noticeable stutters during both gameplay and cut-scenes on Sony's system. The Xbox One game delivers a clean 30fps presentation without these mild inconsistencies, which makes the experience feel a little more refined where performance is concerned.
PS4's so-called "advantage" with MSAA 4X has a penalty i.e. less stable frame rates.
Froza H2 has MSAA 4X with forward plus rendering at 1920x1080p.
Did you forget there's a link between CPU and GPU?
Have you realized CPU's memory consumption floats between 0 to 30 GB/s?
Have you realized the purpose why you have a CPU L2 cache?
Have you realize Pitcairn ES's GDDR5 doesn't require any special programming tricks?
The 7950 BE has better MSAA than the PS4 because it has freaking 28 CU,384 bit bus and higher bandwidth than the PS4 you freaking fool,What the fu** man do you really believe that because a damn GPU has more bandwidth than another that mean it most perform better or equal without taking into notice power.?
GCN's CU has nothing to do with MSAA resolve i.e. it's ROPS unit's MSAA co-processors and memory bandwidth you stupid cow.
Using Radeon HD 4870 as an example which is basically DX10 version of Radeon HD 5770/6770 hardware.
See this is the problem with did i CLAIM no game on xbox one uses MSAA 4X.?
NO i showed you a scenario where the PS4 and xbox one were at even field even resolution wise,1 was MSAA 2X the Other was MSAA 4X so i ask you again which use more bandwidth.
Forza Horizon 2 is a pretty unimpressive game which only impressive feature is MSAA 4X,it look like crap even NFS look better.
The xbox one used 30GB for system you fool it is reserved and no it can't be from 0 to 30GB/s because 0 bandwidth would not run the system why the fu** you think it has 30GB/s.?
MSAA 4X has a performance hit you blind biased fool,so running it on weaker GPU has a cost,so if you have 28 CU vs a GPU that has 18 the one with 28 can sustain that hit better.
Again with No AA we get 69.3 average FPS on the GTX 680. When we turn on just 2X MSAA look at how the framerate drops and the average is now 46.9 FPS. This is just with 2X MSAA on versus No AA. That is an almost 33% performance drop just turning 2X MSAA on! Turning on 4X MSAA reduces performance so much it's only averaging 32.4 FPS. From No AA that is over a 50% drop in performance! You can see how ridiculous the performance drain is for regular MSAA versus FXAA which is only 4.6%.
MSAA 4X has a hit to performance idiot,so if one GPU has much more power it will get a drop in performance but will keep frame higher than the one with a weaker GPU,this is a FACT.
Again you claim the xbox one uses 68GB/s + 104GB/s which is a joke it doesn't and saying from 0 to 30GB/s even worse 0 bandwidth would not run the system,stop using pathetic workaround to beat me,fact is you were wrong period..
That memory math of you...hahahaahahaa
@ronvalencia said:
@tormentos:
For UFC's 1600x900p on XBO/PS4 from http://www.eurogamer.net/articles/digitalfoundry-2014-ea-sports-ufc-face-off
PS4's MSAA 4X vs XBO's MSAA 2X
however, in our demo analysis we found the PS4 code fell behind Xbox One in terms of raw stability, and sadly the situation hasn't changed at all in the final game, despite the availability of a launch-day update
Across different arenas using various combatants, we still see some minor frame-rate drops on PS4 and a few regularly torn frames (limited to the top 10 per cent of the screen, so mostly unnoticeable). The situation has little impact on how the game plays, but the dropped frames do lead to some short but noticeable stutters during both gameplay and cut-scenes on Sony's system. The Xbox One game delivers a clean 30fps presentation without these mild inconsistencies, which makes the experience feel a little more refined where performance is concerned.
PS4's so-called "advantage" with MSAA 4X has a penalty i.e. less stable frame rates.
Hahahaa see this is the problem with so a frame drop it most be MSAA 4X..hahahahaaa
It can't be EA no it as to be MSAA just like on other games it most be the CPU when in reality those games require shit for CPU usage the fact that you have dare argue that Alien Isolation and Dishonored were some how CPU bottleneck on PS4 is a testament to how stupid and biased your arguments are when more CPU intensive games exist on PS4 and they run at higher frames,BF4 is more CPU intensive that Alien Isolation and runs close to 60FPS on PS4 Metro as well...lol
NO i showed you a scenario where the PS4 and xbox one were at even field even resolution wise,1 was MSAA 2X the Other was MSAA 4X so i ask you again which use more bandwidth.
You know the answer and you have not factored the less sustained results. Your argument setup would be valid if PS4 delivered the same smooth/sustained frame rates as XBO's version.
Forza Horizon 2 is a pretty unimpressive game which only impressive feature is MSAA 4X,it look like crap even NFS look better.
That's bachelor of arts argument. "MSAA" is an edge AA that uses depth (i.e. geometry) and color data.
The xbox one used 30GB for system you fool it is reserved and no it can't be from 0 to 30GB/s because 0 bandwidth would not run the system why the fu** you think it has 30GB/s.?
So, what happens to ROPS writes 54 GB/s on DDR3?
What happens if CPU transfers it's command list to GPU and then the GPU actions the command list with coherency turned off?
Have you realized AMD Fusion has non-coherency modes/transfers?
MSAA 4X has a performance hit you blind biased fool,so running it on weaker GPU has a cost,so if you have 28 CU vs a GPU that has 18 the one with 28 can sustain that hit better.
Your argument setup would be valid if PS4 delivered the same smooth/sustained frame rates as XBO's version.
Again you claim the xbox one uses 68GB/s + 104GB/s which is a joke it doesn't and saying from 0 to 30GB/s even worse 0 bandwidth would not run the system,stop using pathetic workaround to beat me,fact is you were wrong period..
Are you claiming XBO can't do PS4's zero CPU bandwidth memory usage?
NO i showed you a scenario where the PS4 and xbox one were at even field even resolution wise,1 was MSAA 2X the Other was MSAA 4X so i ask you again which use more bandwidth.
You know the answer and you have not factored the less sustained results. Your argument setup would be valid if PS4 delivered the same smooth/sustained frame rates as XBO's version.
Forza Horizon 2 is a pretty unimpressive game which only impressive feature is MSAA 4X,it look like crap even NFS look better.
That's bachelor of arts argument. "MSAA" is an edge AA that uses depth (i.e. geometry) and color data.
The xbox one used 30GB for system you fool it is reserved and no it can't be from 0 to 30GB/s because 0 bandwidth would not run the system why the fu** you think it has 30GB/s.?
So, what happens to ROPS writes 54 GB/s on DDR3?
What happens if CPU transfers it's command list to GPU and then the GPU actions the command list with coherency turned off?
Have you realized AMD Fusion has non-coherency modes/transfers?
MSAA 4X has a performance hit you blind biased fool,so running it on weaker GPU has a cost,so if you have 28 CU vs a GPU that has 18 the one with 28 can sustain that hit better.
Your argument setup would be valid if PS4 delivered the same smooth/sustained frame rates as XBO's version.
Again you claim the xbox one uses 68GB/s + 104GB/s which is a joke it doesn't and saying from 0 to 30GB/s even worse 0 bandwidth would not run the system,stop using pathetic workaround to beat me,fact is you were wrong period..
Are you claiming XBO can't do PS4's zero CPU bandwidth memory usage?
Again prove that the few frames it dropped was because of MSAA 4X and not because EA.
You did the same shit with AI and Dishonored which are 2 games that aren't CPU intensive at all.every time the PS4 drops a frame you blame the CPU or not something like MSAA 4X,in fact did you see the video.? It shows it is a bug it drops exactly 1 frame only at times,to 29 which you will not be able to tell from 30,and in fact the PS4 version rise to 31 FPS at once while the XBO remain 30,so your excuse is a joke fact is the PS4 is using more bandwidth period regardless of dropping 1 frame some times.
The game is bland and unimpressive art my ass,is open and bland and is not even close to other racing game quality wise.
Stop you bullshit MS own documents state it 30GB/s reserve for the CPU so your argument is wrong,this is the problem with you,you just can't say i make a mistake no you like a cry baby want to spin things around in order to try to make sense of what doesn't period,the same shit happen with Tile resources which you try to claim the PS4 didn't support tier 3 when it does,when you hold tied to Alien Isolation being CPU intensive and many crappy arguments like those where you know you were wrong...
The frame rate on PS4 is as smooth you idiot,1 frame is nothing this is the kind of shit which make you look like a desperate blind fanboy,so here you stand downplaying Project Cars huge ass gap on PS4 vs XBO,which is as high as 16 FPS but now you want to pretend the xbox one has smoother frames because it runs at 30,while the PS4 runs at 30 and drop to 29..hahahahaa
This is ^^ what make you a HYPOCRITE and the reason your arguments are total bullshit fact is PC PS4 >>>>>>>>>> XBO by as much as 16 FPS and some how you don't think is the xbox one cumbersome memory structure,but some how the PS4 loss a frame and it most be a performance hit because of lack of bandwidth,oh wait even dropping 1 frame the PS4 running 900p 4X MSAA still use more bandwidth than the xbox one at 30 with MSAA 2X.
MSAA x4 at 30 and 29 FPS use more bandwidth than MSAA 2X at 30.
Hahahahaaaaaaaaaaaaaaaaaaaaa.. Now what is your excuse...
The PS4 can't do zero bandwidth usage fool have you ever fu**ing seeing an example chart.?
You see why it show 0 to 140 is a scale it doesn't mean it uses 0 dude what the fu**..Hahaha do you see 176GB/s and the PS4 just reaching 140.? No.?
Yeah then you miss interpret that chart,and judging by how you claim 0 bandwidth because the example show the numbers from 0 to 140 it is clear that your understanding of that chart is null,that chart is from before the PS4 was even launch it is GDC 2013,it was mean to be an example of how the PS4 CPU can eat bandwidth disproportionally.
BUt considering that you fu**ing believe that Alien Isolation is CPU intensive i don't think you ever get it,you are to blind and emotionally invested in defending the crappy xbox one the only thing i can think of is that you are a shill or work for MS some how...lol
Stop you bullshit MS own documents state it 30GB/s reserve for the CPU so your argument is wrong,this is the problem with you,you just can't say i make a mistake no you like a cry baby want to spin things around in order to try to make sense of what doesn't period,the same shit happen with Tile resources which you try to claim the PS4 didn't support tier 3 when it does,when you hold tied to Alien Isolation being CPU intensive and many crappy arguments like those where you know you were wrong...
Again, what happens to 54 GB/s ROPS write for DDR3?
the same shit happen with Tile resources which you try to claim the PS4 didn't support tier 3
Your claim is a misrepresentation. I have stated PS4 supports Tiled resources from GDDR5 between HDD. Tiled resources works better when the machine has more one memory bandwidth tiers. PS4 only has a single memory memory bandwidth tier (single layer GDDR5), hence TMUs only has one fetch rate (i.e. GDDR5) and it can't go any higher than that!!!!
You are arguing without knowledge nor understanding the subject matter.
XBO and gaming PCs are two tier memory bandwidth system.
You did the same shit with AI and Dishonored which are 2 games that aren't CPU intensive at all.every time the PS4 drops a frame you blame the CPU or not something like MSAA 4X,in fact did you see the video.? It shows it is a bug it drops exactly 1 frame only at times,to 29 which you will not be able to tell from 30,and in fact the PS4 version rise to 31 FPS at once while the XBO remain 30,so your excuse is a joke fact is the PS4 is using more bandwidth period regardless of dropping 1 frame some times.
The only shit is your argument setup. Your Alien Isolation example is a "do nothing" scenes i.e. where's the gun fight? where's NPCs? where's the effects?
Eat that.
The frame rate on PS4 is as smooth you idiot,1 frame is nothing this is the kind of shit which make you look like a desperate blind fanboy,so here you stand downplaying Project Cars huge ass gap on PS4 vs XBO,which is as high as 16 FPS but now you want to pretend the xbox one has smoother frames because it runs at 30,while the PS4 runs at 30 and drop to 29..hahahahaa
Not according to digital foundry.
Project Cars again? LOL. SMS PC Lead did state DirectX12 will improve MT scaling for AMD's 8 core CPUs.
MSAA x4 at 30 and 29 FPS use more bandwidth than MSAA 2X at 30.
A single screenshot. Not according to digital foundry. Your argument setup is again flawed.
Again prove that the few frames it dropped was because of MSAA 4X and not because EA.
EA increased MSAA with PS4 build and cause issues.
You see why it show 0 to 140 is a scale it doesn't mean it uses 0 dude what the fu**..Hahaha do you see 176GB/s and the PS4 just reaching 140.? No.?
The only haha is your post since left side bars includes CPU workloads mixed with GPU's workload while right most bar is near 0.
Again, what happens to 54 GB/s ROPS write for DDR3?
the same shit happen with Tile resources which you try to claim the PS4 didn't support tier 3
Your claim is a misrepresentation. I have stated PS4 supports Tiled resources from GDDR5 between HDD. Tiled resources works better when the machine has more one memory bandwidth tiers. PS4 only has a single memory memory bandwidth tier (single layer GDDR5), hence TMUs only has one fetch rate (i.e. GDDR5) and it can't go any higher than that!!!!
You are arguing without knowledge nor understanding the subject matter.
XBO and gaming PCs are two tier memory bandwidth system.
You did the same shit with AI and Dishonored which are 2 games that aren't CPU intensive at all.every time the PS4 drops a frame you blame the CPU or not something like MSAA 4X,in fact did you see the video.? It shows it is a bug it drops exactly 1 frame only at times,to 29 which you will not be able to tell from 30,and in fact the PS4 version rise to 31 FPS at once while the XBO remain 30,so your excuse is a joke fact is the PS4 is using more bandwidth period regardless of dropping 1 frame some times.
The only shit is your argument setup. Your Alien Isolation example is a "do nothing" scenes i.e. where's the gun fight? where's NPCs? where's the effects?
Eat that.
The frame rate on PS4 is as smooth you idiot,1 frame is nothing this is the kind of shit which make you look like a desperate blind fanboy,so here you stand downplaying Project Cars huge ass gap on PS4 vs XBO,which is as high as 16 FPS but now you want to pretend the xbox one has smoother frames because it runs at 30,while the PS4 runs at 30 and drop to 29..hahahahaa
Not according to digital foundry.
Project Cars again? LOL. SMS PC Lead did state DirectX12 will improve MT scaling for AMD's 8 core CPUs.
MSAA x4 at 30 and 29 FPS use more bandwidth than MSAA 2X at 30.
A single screenshot. Not according to digital foundry. Your argument setup is again flawed.
Again prove that the few frames it dropped was because of MSAA 4X and not because EA.
EA increased MSAA with PS4 build and cause issues.
You see why it show 0 to 140 is a scale it doesn't mean it uses 0 dude what the fu**..Hahaha do you see 176GB/s and the PS4 just reaching 140.? No.?
The only haha is your post since left side bars includes CPU workloads mixed with GPU's workload while right most bar is near 0.
1-Who the fu** knows i never claim 54GB/s you did after falsely claiming 68GB/s + 104GB/s,fact is only been able to use 54GB/s is even worse since there is even lower bandwidth and when you split it it has even less than 38GB/s for GPU and is more far apart from your argument of the XBO having Pitcairn like bandwidth,there is no argument about 54GB/s you bring it i don't even know why as a spin.
2-You claimed the PS4 didn't support tier 3,and the other nothing that it work better when it has 2 memory is even worse,as PRT or Tile resources is a GPU features and GPU only have 1 memory structure GDDR5 they don't have ESRAM + GDDR5 PRT is hardware support for virtual texturing is a GCN feature and work on 1 memory address not 2.
3-Dude what the fu** did you even see Alien Isolation benchmark.? It drop more on xbox one than on PS4,there goes your CPU argument the xbox one CPU speed is faster is faster cores is the problem the xbox one should run better,Alien Isolation is not an async shader game..lol
Alien Isolation runs at 104FPS on a i3,is not CPU intensive at all and i quote techspot on it,Techspot >>>>>>>>>> YOU.
Loading Video...
See the video again prove to me that MSAA 2X at 30 FPS requires more bandwidth than MSAA 4X at 29 FPS...lol
Since the argument is bandwidth you should PROVE it,dropping a frame doesn't change the fact that the PS4 uses more bandwidth.
Hahahahaaaaaaaaaaaaaaaaaaaa Ronvalencia ignoring screens proving the PS4 goes even abode the xbox one..hahahaa
PROVE MSAA 2X uses more bandwidth than MSAA 4X even when it is 30 vs 29 the outcome is the same the PS4 was using more bandwidth period,you are a sad biased lemming who don't have the guts to admit been wrong..hahahahaaha
You are a fool a total buffoon,you claim the xbox one and PS4 can ron on 0 bandwidth,which mean they would run without CPU since it is impossible the CPU runs with 0 bandwidth your argument is so pathetic is not even funny,and to think you used the xbox one 30GB/s as advantage vs the PS4 claiming 30GB/s vs 20GB/s hahahaa now you want to pretend it runs at 0 bandwidth because a metric test show results from 0...
Hahahahaaaaaaaaaaaaaaaaaaaaaaaa...
So what is your next argument to justify your stupid claims.? The xbox one runs on love and since you love it more it will outperform the PS4.? Hahahaaaaaaaaaaaa
Nothing new indeed. PS4 remains the world's most powerful video game console in the history of video games.
That's kind of a dumb statement, isn't it? Any console that is the most powerful when a game launches claims that title.
except Xbone. it launched weak af from the get go. sad.
I seem to remember a console called "PlayStation 3" that had trouble matching the "XBox 360" - which came out a year earlier - in multiplat games the first few years... Huh, imagine that.
Again, what happens to 54 GB/s ROPS write for DDR3?
the same shit happen with Tile resources which you try to claim the PS4 didn't support tier 3
Your claim is a misrepresentation. I have stated PS4 supports Tiled resources from GDDR5 between HDD.Tiled resources works better when the machine has more one memory bandwidth tiers. PS4 only has a single memory memory bandwidth tier (single layer GDDR5), hence TMUs only has one fetch rate (i.e. GDDR5) and it can't go any higher than that!!!!
You are arguing without knowledge nor understanding the subject matter.
XBO and gaming PCs are two tier memory bandwidth system.
You did the same shit with AI and Dishonored which are 2 games that aren't CPU intensive at all.every time the PS4 drops a frame you blame the CPU or not something like MSAA 4X,in fact did you see the video.? It shows it is a bug it drops exactly 1 frame only at times,to 29 which you will not be able to tell from 30,and in fact the PS4 version rise to 31 FPS at once while the XBO remain 30,so your excuse is a joke fact is the PS4 is using more bandwidth period regardless of dropping 1 frame some times.
The only shit is your argument setup. Your Alien Isolation example is a "do nothing" scenes i.e. where's the gun fight? where's NPCs? where's the effects?
Eat that.
The frame rate on PS4 is as smooth you idiot,1 frame is nothing this is the kind of shit which make you look like a desperate blind fanboy,so here you stand downplaying Project Cars huge ass gap on PS4 vs XBO,which is as high as 16 FPS but now you want to pretend the xbox one has smoother frames because it runs at 30,while the PS4 runs at 30 and drop to 29..hahahahaa
Not according to digital foundry.
Project Cars again? LOL. SMS PC Lead did state DirectX12 will improve MT scaling for AMD's 8 core CPUs.
MSAA x4 at 30 and 29 FPS use more bandwidth than MSAA 2X at 30.
A single screenshot. Not according to digital foundry. Your argument setup is again flawed.
Again prove that the few frames it dropped was because of MSAA 4X and not because EA.
EA increased MSAA with PS4 build and cause issues.
You see why it show 0 to 140 is a scale it doesn't mean it uses 0 dude what the fu**..Hahaha do you see 176GB/s and the PS4 just reaching 140.? No.?
The only haha is your post since left side bars includes CPU workloads mixed with GPU's workload while right most bar is near 0.
1-Who the fu** knows i never claim 54GB/s you did after falsely claiming 68GB/s + 104GB/s,fact is only been able to use 54GB/s is even worse since there is even lower bandwidth and when you split it it has even less than 38GB/s for GPU and is more far apart from your argument of the XBO having Pitcairn like bandwidth,there is no argument about 54GB/s you bring it i don't even know why as a spin.
2-You claimed the PS4 didn't support tier 3,and the other nothing that it work better when it has 2 memory is even worse,as PRT or Tile resources is a GPU features and GPU only have 1 memory structure GDDR5 they don't have ESRAM + GDDR5 PRT is hardware support for virtual texturing is a GCN feature and work on 1 memory address not 2.
3-Dude what the fu** did you even see Alien Isolation benchmark.? It drop more on xbox one than on PS4,there goes your CPU argument the xbox one CPU speed is faster is faster cores is the problem the xbox one should run better,Alien Isolation is not an async shader game..lol
Alien Isolation runs at 104FPS on a i3,is not CPU intensive at all and i quote techspot on it,Techspot >>>>>>>>>> YOU.
See the video again prove to me that MSAA 2X at 30 FPS requires more bandwidth than MSAA 4X at 29 FPS...lol
Since the argument is bandwidth you should PROVE it,dropping a frame doesn't change the fact that the PS4 uses more bandwidth.
Hahahahaaaaaaaaaaaaaaaaaaaa Ronvalencia ignoring screens proving the PS4 goes even abode the xbox one..hahahaa
PROVE MSAA 2X uses more bandwidth than MSAA 4X even when it is 30 vs 29 the outcome is the same the PS4 was using more bandwidth period,you are a sad biased lemming who don't have the guts to admit been wrong..hahahahaaha
You are a fool a total buffoon,you claim the xbox one and PS4 can ron on 0 bandwidth,which mean they would run without CPU since it is impossible the CPU runs with 0 bandwidth your argument is so pathetic is not even funny,and to think you used the xbox one 30GB/s as advantage vs the PS4 claiming 30GB/s vs 20GB/s hahahaa now you want to pretend it runs at 0 bandwidth because a metric test show results from 0...
Hahahahaaaaaaaaaaaaaaaaaaaaaaaa...
So what is your next argument to justify your stupid claims.? The xbox one runs on love and since you love it more it will outperform the PS4.? Hahahaaaaaaaaaaaa
1. Your fix "38 GB/s" is wrong e.g. if you notice PS4's version, CPU allocation changes from near 0 to about 10 GB/s.
AMD PC APUs doesn't operate in fix CPU/GPU bandwidth allocations.
Prove Pitcairn at 153.6 GB/s memory bandwidth at 100 percent efficiency. I have shown you 5870's GDDR5 efficiency at ~72 percent efficiency.
If you apply 72 percent efficiency for PS4's 176 GB/s, it yields 126.5 GB/s which is very close Sony's own numbers i.e. ~130 GB/s.
Like Pitcairn ES/Pro/XT's 153.6 GB/s, 5870's also has 153 GB/s GDDR5 memory operations.
A non-fanboy view would apply the same efficiency standard across all AMD GPU SKUs.
Prior to AMD Tonga, NVIDIA GPUs has better color compression.
2. Wrong, PS4's supports Tile resources per as AMD's PC APUs with GCN IGPs, but the gain is very minor since PS4 only has a single memory bandwidth tier i.e. texture fetch rates remains the same i.e. 176 GB/s against efficiency percentage.
3. Alien Isolation XBO vs PS4 is not my argument. My argument for Alien Isolation is PC's 7770 result is superior to both XBO and PS4. Hint: it's a Master Race win.
As an AMD Gaming Evolved title, Alien Isolation PC doesn't run into DX11 draw calls limitations i.e. a known AMD PC DX11 driver problem.
Alien Isolation PC's recommended specs from http://store.steampowered.com/app/214490
A direct port to console's quad-cores at 1.6Ghz or 1.7Ghz would be not in PC's recommended specs.
Intel Core i series vs AMD FX and Phenom II
Intel Core i7 with dual core+HT (4 threads) mode rivals AMD Phenom II with quad core.
Intel Core i3-2100 (Dual core/4 threads at 3.1Ghz) ~= Intel Core 2 Quad Q9400 (Quad core/4 threads at 2.66Ghz) ~= AMD A10-5800K (Quad core/4 threads at 3.8Ghz) just under Phenom II X4 840 (Quad Core/4 threads at 3.2Ghz).
PROVE MSAA 2X uses more bandwidth than MSAA 4X even when it is 30 vs 29 the outcome is the same the PS4 was using more bandwidth period,you are a sad biased lemming who don't have the guts to admit been wrong..hahahahaaha
I don't need to PROVE anything since I didn't made "MSAA 2X uses more bandwidth than MSAA 4X" claim, but Forza H2 has MSAA 4X (with forward rendering).
Increasing MSAA levels impacts two factors i.e. data storage and memory bandwidth.
Both of these machines are still using rendering techniques and rendering 'cheats' that were intended for previous console based architecture.
Once everything gets moved over to compute the gap between the 2 machines will become more apparent as there's so much that's still done on CPU that can be done a lot faster via compute.
Both of these machines are still using rendering techniques and rendering 'cheats' that were intended for previous console based architecture.
Once everything gets moved over to compute the gap between the 2 machines will become more apparent as there's so much that's still done on CPU that can be done a lot faster via compute.
It depends if XBO's two ACE units with 8 queues at 853Mhz each are enough to saturate the it's 12 CUs. If both console GPUs are saturated, the gap will remain the same i.e XBO is 70 percent of PS4 in CU power, but there are operations that memory bandwidth bound.
Both ACE and GCP units can process compute workloads.
For example, 48 TMUs was able to improve XBO's frame buffer BW usage by 70 to 80 percent, hence running into memory bandwidth bound.
XBO's two ACE units vs 12 CU ratio is similar to Radeon HD R9-290X's two ACE units vs 11 CU ratio.
PS4 has an advantage in TMU fetch (read) rates since it's primary texture memory pool has higher memory bandwidth i.e. PS4's ~130 GB/s (based from Sony's numbers without CPU access) vs XBO's ~53 GB/s, (via DDR3) but texture reads from XBO's ESRAM would be memory bandwidth bound.
PS4 has an advantage in TMU filtering processor count i.e. PS4's 72 at 800Mhz vs XBO's 48 at 853Mhz i.e. XBO is 70 percent of PS4.
XBO has a slight advantage in tessellation/geometry i.e. PS4 is 93 percent of XBO.
XBO has an advantage in CPU power for feeding the ACE units i.e. PS4 is ~86 percent of XBO.
1. Your fix "38 GB/s" is wrong e.g. if you notice PS4's version, CPU allocation changes from near 0 to about 10 GB/s.
AMD PC APUs doesn't operate in fix CPU/GPU bandwidth allocations.
Prove Pitcairn at 153.6 GB/s memory bandwidth at 100 percent efficiency. I have shown you 5870's GDDR5 efficiency at ~72 percent efficiency.
If you apply 72 percent efficiency for PS4's 176 GB/s, it yields 126.5 GB/s which is very close Sony's own numbers i.e. ~130 GB/s.
Like Pitcairn ES/Pro/XT's 153.6 GB/s, 5870's also has 153 GB/s GDDR5 memory operations.
A non-fanboy view would apply the same efficiency standard across all AMD GPU SKUs.
Prior to AMD Tonga, NVIDIA GPUs has better color compression.
2. Wrong, PS4's supports Tile resources per as AMD's PC APUs with GCN IGPs, but the gain is very minor since PS4 only has a single memory bandwidth tier i.e. texture fetch rates remains the same i.e. 176 GB/s against efficiency percentage.
3. Alien Isolation XBO vs PS4 is not my argument. My argument for Alien Isolation is PC's 7770 result is superior to both XBO and PS4. Hint: it's a Master Race win.
As an AMD Gaming Evolved title, Alien Isolation PC doesn't run into DX11 draw calls limitations i.e. a known AMD PC DX11 driver problem.
Alien Isolation PC's recommended specs from http://store.steampowered.com/app/214490
A direct port to console's quad-cores at 1.6Ghz or 1.7Ghz would be not in PC's recommended specs.
Intel Core i series vs AMD FX and Phenom II
Intel Core i7 with dual core+HT (4 threads) mode rivals AMD Phenom II with quad core.
Intel Core i3-2100 (Dual core/4 threads at 3.1Ghz) ~= Intel Core 2 Quad Q9400 (Quad core/4 threads at 2.66Ghz) ~= AMD A10-5800K (Quad core/4 threads at 3.8Ghz) just under Phenom II X4 840 (Quad Core/4 threads at 3.2Ghz).
PROVE MSAA 2X uses more bandwidth than MSAA 4X even when it is 30 vs 29 the outcome is the same the PS4 was using more bandwidth period,you are a sad biased lemming who don't have the guts to admit been wrong..hahahahaaha
I don't need to PROVE anything since I didn't made "MSAA 2X uses more bandwidth than MSAA 4X" claim, but Forza H2 has MSAA 4X (with forward rendering).
Increasing MSAA levels impacts two factors i.e. data storage and memory bandwidth.
1-No is not wrong you blind biased lemming.
The xbox ones has 68GB/s peak from that pool of Ram,the XBO has 30GB/s lock for the CPU,what is left from the peak is 38Gbs/s the xbox one GPU can't use all 68GB/s because that would have starve the system,why the fu** you think ESRAM is there in the first place.?
Is there because 68GB/s wasn't enough for a GPU like the 7770 which has 72GB's alone for it,and its CPU which on PC tend to have 68GB/s for self.
So your 68GB/s + 104 GB/s is WRONG and you should admit it,there is no spin you can use you were wrong,just like claiming AVX support for G1610.
The PS4 is not a PC it has an APU with GDDR5 as system memory as well it doesn't have the same memory controllers as a PC which has DDR3 and GDDR5,nor work on the same way,how many of AMD APU use GDDR5.? Because the A10 and other use system memory DDR3.
2-All GCN have tier 3 tile resources,and all of them use 1 memory structure on their GPU not like the xbox one,no GCN from AMD has ESRAM so stop your silly ass argument it doesn't need 2 memory pools you fool.
Loading Video...
That shit was already debunked Trials Fusion is a virtual texture game and runs more consistent on PS4 and at higher resolution to,that shit was debunked more than a year ago,just like the silly notion of GDDR5 stalling the PS4 CPU..lol
3-You argument about Alien Isolation is about the PS4 CPU and xbox one not been enough for the game,when in reality that game isn't CPU intensive.
This is why i consider you a blind fool you most have problems mentally because no one is as dumb to own it self again an again on such thing,Project Cars requires a damn i7 CPU and the PS4 runs it at close to 60FPS with 20 cars with 45 still runs it like 40 FPS faster than Alien Isolation in fact Project Cars CPU requirement MINIMUM is higher than Alien Isolation Recommended specs so your whole shitty argument is a joke.
Recommended:
OS: Windows 7 64-Bit / Windows 8 64-Bit / Windows 8.1 64-Bit
COD Advanced Warfare CPU requirements oh look it requires a i5 2500k 3.3ghz Is RECOMMENDS an i5 2500k at 3.3 ghz that CPU own the Phenom II X4 955 - 4 Core, 3.2 GHz.
Loading Video...
I just chewed and spit your whole argument about the game needing Phenom II X4 955 - 4 Core, 3.2 GHz which is the recommended specs,COD Advanced Warfare requires more CPU and is 60FPS 1080p on PS4 by your logic this games should not run at 60FPS because its recommended CPU specs are over the PS4 ones.
Oh and by the way before you hide on a direct port excuse, a direct port would indicate that the spec of the consoles were not taken into notice and that no further optimization was done to redistribute all load around the consoles 6 CPU cores,which MEAN a butcher job or been lazy which is what i have being saying all alone this game was.
Thank you for admitting it...lol
You didn't claim it did,you are just downplaying the comparison based on the PS4 dropping a single frame,so i tell you is irrelevant because MSAA 4X even at 29FPS uses more Bandwidth than MSAA xa at 30 and you know it..lol
Yeah it uses MSAA 4X but the graphics look like shit is a bland looking game and certaining not in the same level as the Order visually.
@ronvalencia said:
It depends if XBO's two ACE units with 8 queues at 853Mhz each are enough to saturate the it's 12 CUs. If both console GPUs are saturated, the gap will remain the same i.e XBO is 70 percent of PS4 in CU power, but there are operations that memory bandwidth bound.
Both ACE and GCP units can process compute workloads.
For example, 48 TMUs was able to improve XBO's frame buffer BW usage by 70 to 80 percent, hence running into memory bandwidth bound.
XBO's two ACE units vs 12 CU ratio is similar to Radeon HD R9-290X's two ACE units vs 11 CU ratio.
PS4 has an advantage in TMU fetch (read) rates since it's primary texture memory pool has higher memory bandwidth i.e. PS4's ~130 GB/s (based from Sony's numbers without CPU access) vs XBO's ~53 GB/s, (via DDR3) but texture reads from XBO's ESRAM would be memory bandwidth bound.
PS4 has an advantage in TMU filtering processor count i.e. PS4's 72 at 800Mhz vs XBO's 48 at 853Mhz i.e. XBO is 70 percent of PS4.
XBO has a slight advantage in tessellation/geometry i.e. PS4 is 93 percent of XBO.
XBO has an advantage in CPU power for feeding the ACE units i.e. PS4 is ~86 percent of XBO.
Hahahaaaaaaaaaaaaaaaaaa... See so DX12 will fix the xbox one problems but more Aces on PS4 will do nothing... And you are not a Fanboy right.?
So sony chose 8 Aces for laughs and giggles,now where did i see some derange argument like this before...????
Oh shit it was Major damage control nelson talking shit about Cell and how it wasn't good for gaming...Hahahaa
The Cell’s seven DSPs (what Sony calls SPEs) have no cache, no direct access to memory, no branch predictor, and a different instruction set from the PS3’s main CPU. They are not designed for or efficient at general purpose computing. DSPs are not appropriate for game programming.
Xbox 360 has three general purpose CPU cores. The Cell processor has only one.
Xbox 360’s CPUs has vector processing power on each CPU core. Each Xbox 360 core has 128 vector registers per hardware thread, with a dot product instruction, and a shared 1-MB L2 cache. The Cell processor’s vector processing power is mostly on the seven DSPs.
Sony’s CPU is ideal for an environment where 12.5% of the work is general-purpose computing and 87.5% of the work is DSP calculations. That sort of mix makes sense for video playback or networked waveform analysis, but not for games.
Hahahahaaaaaaaaaaaaa Major Nelson stating Cell wasn't good for games...hahahahahaa
The end result the Cell owning the Xenon by 3 to 1 margin using 5SPE with 6 it would have beat the xbox one results as well.lol
So dude i am sure sony knew what they were doing when they put those 8 Aces there i am more than sure,last time people say shit about Cell even Major Nelson it self and he was wrong,and Cell ended giving the edge powerwise to the PS3..
The xbox ones has 68GB/s peak from that pool of Ram,the XBO has 30GB/s lock for the CPU,what is left from the peak is 38Gbs/s the xbox one GPU can't use all 68GB/s because that would have starve the system,why the fu** you think ESRAM is there in the first place.?(1)
Is there because 68GB/s wasn't enough for a GPU like the 7770 which has 72GB's alone for it,and its CPU which on PC tend to have 68GB/s for self.
So your 68GB/s + 104 GB/s is WRONG and you should admit it,there is no spin you can use you were wrong,just like claiming AVX support for G1610.
The PS4 is not a PC it has an APU with GDDR5 as system memory as well it doesn't have the same memory controllers as a PC which has DDR3 and GDDR5,nor work on the same way,how many of AMD APU use GDDR5.? Because the A10 and other use system memory DDR3.
1. 30 GB/s is not locked by CPU. Why the fu** you think CPU cache is there in the first place? Hint: reduce main memory access hit rates.
AMD Jaguar quad core's 2MB L2 cache size is 2X larger than PPE X3(6 threads)'s 1MB L2 cache. Both XBO and PS4 has two quad-core Jaguar modules with a total of 4MB L2 cache which is 4X over Xbox 360's 1MB L2 cache.
Sony's reveal on PS4's practical memory bandwidth usage shows typical AMD memory controller efficiencies. Sony > YOU.
@tormentos:
2-All GCN have tier 3 tile resources,and all of them use 1 memory structure on their GPU not like the xbox one,no GCN from AMD has ESRAM so stop your silly ass argument it doesn't need 2 memory pools you fool.
I have shown you PC OpenGL example that used main memory(slow memory) to GDDR5 (fast memory) for it's PRT. Trials Fusion use PRT to stream-in textures from hard drive i.e. to shorten the loading time.
@tormentos:
That shit was already debunked Trials Fusion is a virtual texture game and runs more consistent on PS4 and at higher resolution to,that shit was debunked more than a year ago,just like the silly notion of GDDR5 stalling the PS4 CPU..lol
You don't know what you are talking about. If PRT was applied for hard drive streaming into memory, the bottleneck is the hard drive.
These streaming issues are especially glaring on PS4 during the Turbine Terror stage, where the blurred texture beneath our tyres fails to update until we're far past it. Pop-in also flares up aggressively when restarting from various checkpoints on Xbox One, and here even shadow maps can be found snapping into view. In the reference PC version, however, pop-in goes almost unnoticed if you're playing on a machine with an SSD to stream from.
You are making an argument without knowledge.
@tormentos:
3-You argument about Alien Isolation is about the PS4 CPU and xbox one not been enough for the game,when in reality that game isn't CPU intensive.
This is why i consider you a blind fool you most have problems mentally because no one is as dumb to own it self again an again on such thing,Project Cars requires a damn i7 CPU and the PS4 runs it at close to 60FPS with 20 cars with 45 still runs it like 40 FPS faster than Alien Isolation in fact Project Cars CPU requirement MINIMUM is higher than Alien Isolation Recommended specs so your whole shitty argument is a joke.
For Project Cars PC and four threads, i3's single thread performance will be reduced when compared with i5's full speed four threads AND PC version is known have draw call bottleneck from single thread rendering. PC patches has focused on reducing the draw calls to increase the frame rates. Project Cars PC version didn't follow PS4's four thread rendering i.e. remember, devs has stated PS4's four thread rendering = high end PC CPU's single thread rendering. Intel i3 with quad threads would break that hardware assumption.
Alien Isolation PC version is not known to have draw call bottlenecks, hence i3 with four threads is able run efficiently. For PC, badly using batching workaround will cause excessive draw calls cost. Without Direct3D getting in a way, Intel Core I3-2100 rivals 3 Ghz AMD quad core/4 thread solutions and my pure CPU benchmarks are properly mufti-threaded. Your Project Cars PC (Intel Core i3) vs consoles argument is obsolete with the confirmed DirectX12 build.
Again, you are making an argument without knowledge.
Your argument on low end PC CPU vs console CPU is irrelevant since PC has the option to upgrade it's CPU. The focus is console's GPU under preforming results relative PC's 7770 result regardless of PC CPU.
You didn't claim it did,you are just downplaying the comparison based on the PS4 dropping a single frame,so i tell you is irrelevant because MSAA 4X even at 29FPS uses more Bandwidth than MSAA xa at 30 and you know it..lol
Digital Foundry has stated and has focused on "cut scenes.".
Of course MSAA 4X will consume extra storage space and memory bandwidth when compared to MSAA 2X.
It's EA's decision to cause PS4 version to have frame tearing results by slightly over taxing the PS4 e.g. MSAA 4X.
@tormentos said:
Yeah it uses MSAA 4X but the graphics look like shit is a bland looking game and certaining not in the same level as the Order visually.
The problem with your statement is artwork subjectivity. The Order's 1920x800p was gimped by memory bandwidth i.e. access depth data and applying edge AA on frame buffer color data.
1600x900p (1,440,000 pixels) MSAA 4X is close to 1920x800p (1,536,000 pixel) MSAA 4X.
@tormentos said:
Hahahaaaaaaaaaaaaaaaaaa... See so DX12 will fix the xbox one problems but more Aces on PS4 will do nothing... And you are not a Fanboy right.?
The reason,
R9-290X has 8 ACE units with 44 CUs i.e. the ratio is 2 ACE against 11 CU.
R9-380X has 8 ACE units with 32 CUs i.e. the ratio is 2 ACE against 4 CU.
Oh shit it was Major damage control nelson talking shit about Cell and how it wasn't good for gaming...Hahahaa
The difference, AMD ACE units operates within the GPU not out side of it as in the case of PS3's setup.
AMD ACE units has direct access to GPU's hardware you stupid cow.
So dude i am sure sony knew what they were doing when they put those 8 Aces there i am more than sure,last time people say shit about Cell even Major Nelson it self and he was wrong,and Cell ended giving the edge powerwise to the PS3..
SPE doesn't have direct access to GPU hardware you stupid cow.
You didn't claim it did,you are just downplaying the comparison based on the PS4 dropping a single frame,so i tell you is irrelevant because MSAA 4X even at 29FPS uses more Bandwidth than MSAA xa at 30 and you know it..lol
Digital Foundry has stated and has focused on "cut scenes.".
Of course MSAA 4X will consume extra storage space and memory bandwidth when compared to MSAA 2X.
It's EA's decision to cause PS4 version to have frame tearing results by slightly over taxing the PS4 e.g. MSAA 4X.
@tormentos said:
Yeah it uses MSAA 4X but the graphics look like shit is a bland looking game and certaining not in the same level as the Order visually.
The problem with your statement is artwork subjectivity. The Order's 1920x800p was gimped by memory bandwidth i.e. access depth data and applying edge AA on frame buffer color data.
1600x900p (1,440,000 pixels) MSAA 4X is close to 1920x800p (1,536,000 pixel) MSAA 4X.
@tormentos said:
Hahahaaaaaaaaaaaaaaaaaa... See so DX12 will fix the xbox one problems but more Aces on PS4 will do nothing... And you are not a Fanboy right.?
The reason,
R9-290X has 8 ACE units with 44 CUs i.e. the ratio is 2 ACE against 11 CU.
R9-380X has 8 ACE units with 32 CUs i.e. the ratio is 2 ACE against 4 CU.
Oh shit it was Major damage control nelson talking shit about Cell and how it wasn't good for gaming...Hahahaa
The difference, AMD ACE units operates within the GPU not out side of it as in the case of PS3's setup.
AMD ACE units has direct access to GPU's hardware you stupid cow.
So dude i am sure sony knew what they were doing when they put those 8 Aces there i am more than sure,last time people say shit about Cell even Major Nelson it self and he was wrong,and Cell ended giving the edge powerwise to the PS3..
SPE doesn't have direct access to GPU hardware you stupid cow.
That is what i have claim period.
The order is not 1920x800p because of MSAA,it because they wanted a cinematic experience,the game not delivering MSAA X4 at 1080p was do to power not just bandwidth alone,more pixels with higher AA means lower frames.
Is not about the fu**ing ratio for got sake Mode queues faster completion period,did AMD state it doesn't apply to the PS4.? No.? Then you have shit,unless you have proof of the PS4 using its 8 Aces vs the xbox one 2 and having no gain you have nothing the xbox one or PS4 are not PC period,the same shit Major Nelson claimed about Cell SPE and ended eating up his words one by one..hahaha
It doesn't matter is a way to feed the extra CU in the PS4 more jobs,the PS4 has 18 CU you know not 12 like the xbox one,and GPU are composed of hundreds of cores so debide 1152 SP over 64 queues.
And you will see the PS4 has more queues per CU than the xbox one,which mean it has more ways to summit jobs to those 1152 SP.
The way Async Shaders work is like a toll with many lines were ever one is free the access dispatch a job over the free line.
There is no downside to this and probably explain why Ubisoft compute test was so one sided for the PS4 GPU.
But more access is a way to summit more jobs to the hundreds of SP inside CU you,blind lemming....
You didn't claim it did,you are just downplaying the comparison based on the PS4 dropping a single frame,so i tell you is irrelevant because MSAA 4X even at 29FPS uses more Bandwidth than MSAA xa at 30 and you know it..lol
Digital Foundry has stated and has focused on "cut scenes.".
Of course MSAA 4X will consume extra storage space and memory bandwidth when compared to MSAA 2X.
It's EA's decision to cause PS4 version to have frame tearing results by slightly over taxing the PS4 e.g. MSAA 4X.
@tormentos said:
Yeah it uses MSAA 4X but the graphics look like shit is a bland looking game and certaining not in the same level as the Order visually.
The problem with your statement is artwork subjectivity. The Order's 1920x800p was gimped by memory bandwidth i.e. access depth data and applying edge AA on frame buffer color data.
1600x900p (1,440,000 pixels) MSAA 4X is close to 1920x800p (1,536,000 pixel) MSAA 4X.
@tormentos said:
Hahahaaaaaaaaaaaaaaaaaa... See so DX12 will fix the xbox one problems but more Aces on PS4 will do nothing... And you are not a Fanboy right.?
The reason,
R9-290X has 8 ACE units with 44 CUs i.e. the ratio is 2 ACE against 11 CU.
R9-380X has 8 ACE units with 32 CUs i.e. the ratio is 2 ACE against 4 CU.
Oh shit it was Major damage control nelson talking shit about Cell and how it wasn't good for gaming...Hahahaa
The difference, AMD ACE units operates within the GPU not out side of it as in the case of PS3's setup.
AMD ACE units has direct access to GPU's hardware you stupid cow.
So dude i am sure sony knew what they were doing when they put those 8 Aces there i am more than sure,last time people say shit about Cell even Major Nelson it self and he was wrong,and Cell ended giving the edge powerwise to the PS3..
SPE doesn't have direct access to GPU hardware you stupid cow.
That is what i have claim period.
The order is not 1920x800p because of MSAA,it because they wanted a cinematic experience,the game not delivering MSAA X4 at 1080p was do to power not just bandwidth alone,more pixels with higher AA means lower frames.
Is not about the fu**ing ratio for got sake Mode queues faster completion period,did AMD state it doesn't apply to the PS4.? No.? Then you have shit,unless you have proof of the PS4 using its 8 Aces vs the xbox one 2 and having no gain you have nothing the xbox one or PS4 are not PC period,the same shit Major Nelson claimed about Cell SPE and ended eating up his words one by one..hahaha
It doesn't matter is a way to feed the extra CU in the PS4 more jobs,the PS4 has 18 CU you know not 12 like the xbox one,and GPU are composed of hundreds of cores so debide 1152 SP over 64 queues.
And you will see the PS4 has more queues per CU than the xbox one,which mean it has more ways to summit jobs to those 1152 SP.
The way Async Shaders work is like a toll with many lines were ever one is free the access dispatch a job over the free line.
There is no downside to this and probably explain why Ubisoft compute test was so one sided for the PS4 GPU.
But more access is a way to summit more jobs to the hundreds of SP inside CU you,blind lemming....
The devs has stated it's either MSAA 2X + 1920x1080p or MSAA 4X + 1920x800p. "Cinematic experience" is just an excuse to have black bars. Where's your PS4 1920x1080p with MSAA 4X example blind cow?
Submitting more jobs to a saturated GPU doesn't increase fps performance you stupid cow.
ACE is a device to off-load queue management for the CPU side.
Log in to comment