The arguments against a shared Xbox One and PC ecosystem are quite silly

Avatar image for Antwan3K
Antwan3K

9320

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#251  Edited By Antwan3K
Member since 2005 • 9320 Posts

Well, it has been revealed that Quantum Break (and the rumored Gears 4, Forza 6, etc) will all be exclusives to the Windows Store and will not be released on Steam..

Which once again proves my point that this is all about combining and leveraging the PC and Xbox One platforms under one unified "Windows 10" banner..

All you PS4 fanboys that hate Microsoft so much, have fun building your Windows 10 gaming PCs, complete with DirectX12, in order to buy these titles via the Windows Store with Xbox Live achievements, cross-platform party chat, cross-buy, cross-save, and cross-play with the Xbox One console..

Thanks for your support and welcome to the Xbox Live family!!!.. See ya online! :-)

Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#252  Edited By ronvalencia
Member since 2008 • 29612 Posts

@Antwan3K said:

Let's start with the elephant in the room: The announcement that Quantum Break is coming to Windows 10 the same day as Xbox One..

if you want Quantum Break, you still can't play this game on PS4.. You either have to buy/own a $350 Xbox One or buy/build/maintain a potentially $1350-$2000+ gaming PC with Windows 10 and DX12.. If you happen to own both, you get the added benefit of cross-buy and cross-save so that your gaming content and data floats from device to device seamlessly..

...

Windows Store+DirectX12+Sandbox ecosystem effectively turns Windows 10 PC into Xbox One console but with better hardware (GPU greater than R7-260 non-X).

Windows 10 PC clones turned into games console is similar to 3DO game console business model.

Xbox One wouldn't be required if OEM PC vendors that majority ships with Intel IGP matches the minimum R7-260 level GPU i.e. Xbox sets the minimum desktop gaming GPU level for Microsoft's desktop platforms.

The main reason why Microsoft entered hardware business such as Surface/Surface Pro tablet and Surface Book is that Microsoft can't wait for PC OEMs to move to a certain quality level.

Intel is not using it's 14 nm process tech to scale it's IGP to match R7-260 i.e. Intel is more focus on larger profit margins with barely good enough IGP i.e. it keeps the chip size relatively small.

Microsoft is attempting to create interactive content player device standard i.e. start another VHS vs Beta max clone wars but for interactive content player device.

AMD is promising that their new Polaris (GCN 4.0) for "thin and light" notebook will deliver current gen console GPU class power i.e. current AMD FX-8800p SoC for tablet hybrid solution doesn't deliver R7-260 or R7-265 level GPU.

The majority Windows PC form factor is "thin and light" laptops. DTR gaming laptop bricks are in the minority.

Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#253  Edited By ronvalencia
Member since 2008 • 29612 Posts

@GarGx1 said:

What the blazes does the mobile market have to do with Xbox One and PC? Besides how can they lose what they've never had in the first place?

The Asian market, especially China, is the largest single gaming market in the world today

Source

You know there is a reason why both Sony and Microsoft have fought long and hard to get their consoles into China. They want a slice of that $22 Billion, the majority of which is gathered on the PC environment. Console games haven't sold well there because they have never managed to penetrate the market, putting them on PC is one possible way of garnering interest.

What you're failing to understand here is that I'm not saying it is what MS are up to but a POSSIBLE outcome and goal. No one really knows but MS and it's more feasible than MS abandoning the Xbox brand. On that note you, as a Sony fan, really don't want MS to abandon console gaming because you really don't want a gaming environment controlled by a near monopoly. (hint: Monopolies are very bad)

Is a 5 year out of date chart made by a Nintendo fanboy the best you can find for evidence that MS have spent billions chasing Sony? Come on you can do better than that!

If the best version of multi-plat games as well as getting big 'console exclusives' (SFV, Quantum Break etc.) are scraps then I'll happily keep them coming to augment the vast amount of PC exclusives that we get and never appear on console.

You haven't factored in single market areas such as EU or NAFDA (USA, Canada, Mexico).

Australia and New Zealand has EU style CER single market zone.

Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#254  Edited By ronvalencia
Member since 2008 • 29612 Posts

@tormentos said:

@tdkmillsy:

A 7770 would beat the xbox one on this game i am sure.

Is not that expensive

Not when a game's geometry assets is designed for XBO's two triangles per cycle at 853Mhz limits. 7770/R7-250X is only 1 triangle per cycle at 1Ghz.

Quantum Break doesn't have any assumptions for Xbox 360 (1 triangle per cycle at 500Mhz)'s geometry assets. Quantum Break PC's minimum PC GPU is R7-260X(two triangles per cycle) and FX-6300 (6 hardware threads).

Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#255 ronvalencia
Member since 2008 • 29612 Posts

@leandrro said:

@Antwan3K: "buy/build/maintain a potentially $1350-$2000+ gaming PC" lol a $300 PC will run it as good as the X1, the requirements (for a much better quality than X1) is a fx-6300 + r7-260x

PCPartPicker part list: http://pcpartpicker.com/p/FYTVhM

Price breakdown by merchant: http://pcpartpicker.com/p/FYTVhM/by_merchant/

CPU: AMD FX-6300 3.5GHz 6-Core Processor ($98.89 @ OutletPC)

Motherboard: MSI 760GMA-P34(FX) Micro ATX AM3+ Motherboard ($49.95 @ OutletPC)

Memory: PNY Anarchy 8GB (2 x 4GB) DDR3-2133 Memory ($35.99 @ Amazon)

Storage: Western Digital AV-GP 500GB 3.5" 5400RPM Internal Hard Drive ($35.90 @ Amazon)

Video Card: Gigabyte Radeon R7 360 2GB Video Card ($94.99 @ Micro Center)

Case: Enermax Thorex ATX Mid Tower Case ($19.99 @ Newegg)

Power Supply: EVGA 430W 80+ Certified ATX Power Supply ($23.98 @ Newegg)

Operating System: Microsoft Windows 10 Home OEM (64-bit) ($87.95 @ OutletPC)

Total: $447.64

Prices include shipping, taxes, and discounts when available

Generated by PCPartPicker 2016-02-15 20:52 EST-0500

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#256 tormentos
Member since 2003 • 33793 Posts

@ronvalencia said:

Not when a game's geometry assets is designed for XBO's two triangles per cycle at 853Mhz limits. 7770/R7-250X is only 1 triangle per cycle at 1Ghz.

Quantum Break doesn't have any assumptions for Xbox 360 (1 triangle per cycle at 500Mhz)'s geometry assets. Quantum Break PC's minimum PC GPU is R7-260X(two triangles per cycle) and FX-6300 (6 hardware threads).

Pushing more triangle when you are not able to shade them is a problem,which is what CD Project stated with DX12 and the xbox one situation.

Stop using minimum specs you from all the people YOU should now that minimum specs mean nothing,in reality PC with lower specs than those will run the game and probably in better form than the xbox one,the xbox one doesn't have magic on it so that it can perform magically better than other AMD GPU.

And since we all know that all your predictions about DX12,tile resources,esram and all that crap have all turn into salt,i say there is a pretty good chance again you end wrong,hell you can even OC your 7770 which you can't do with the xbox one.

Fact is the 7770 has constantly outperform the xbox one i don't see why it can't do it again.

Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#257  Edited By ronvalencia
Member since 2008 • 29612 Posts

@tormentos said:
@ronvalencia said:

Not when a game's geometry assets is designed for XBO's two triangles per cycle at 853Mhz limits. 7770/R7-250X is only 1 triangle per cycle at 1Ghz.

Quantum Break doesn't have any assumptions for Xbox 360 (1 triangle per cycle at 500Mhz)'s geometry assets. Quantum Break PC's minimum PC GPU is R7-260X(two triangles per cycle) and FX-6300 (6 hardware threads).

Pushing more triangle when you are not able to shade them is a problem,which is what CD Project stated with DX12 and the xbox one situation.

Stop using minimum specs you from all the people YOU should now that minimum specs mean nothing,in reality PC with lower specs than those will run the game and probably in better form than the xbox one,the xbox one doesn't have magic on it so that it can perform magically better than other AMD GPU.

And since we all know that all your predictions about DX12,tile resources,esram and all that crap have all turn into salt,i say there is a pretty good chance again you end wrong,hell you can even OC your 7770 which you can't do with the xbox one.

Fact is the 7770 has constantly outperform the xbox one i don't see why it can't do it again.

Tell that NVIDIA's mobile Maxwellv1 GPUs.

The fact is PC's 7770 is usually backed by stronger Intel CPU just like my laptop with Intel Core i7-3635QM CPU (2.4 Ghz with 3.4Ghz Turbo) and Radeon HD 8870M (7770 based GCN) graphics card.

My point with PC's minimum specs is about obtaining similar performance experience between PC hardware and counter your i3 arguments. Your i3 arguments are not applicable for DX12 Quantum Break.

There's "Screen Space" based shading LOL. Unless you have massive shading overdraw issues, you only shade the polygon surfaces that are visible. Deferred shading is a screen-space shading. Screen-Space Ambient Occlusion is another screen space shading, hence the part of it's name.

Crack Down 3's large scale destruction sequence has shown what XBO's GPU can do when it's geometry dot computation was done via remote Intel Xeon (iCore series). Intel Xeon servers generates the geometry dot destruction sequence for XBO's GPU to play back i.e. from XBO's POV, it's like playing back pre-baked destruction sequence, but the this pre-baked destruction sequence was real time generated by Intel Xeon servers.

As for XBO's tiled resource and ESRAM..

From http://www.gamepur.com/news/16560-dev-explains-forza-horizon-2-1080p4x-msaa-management-says-controversial.html

For FH2, textures partially resident (PRT) was applied for ESRAM, hence maximising TMU's reads/write hence XBO is not limited to 7770's 72 GB/s like TMU read/write memory behavior.

I'm still wait for PS4's 1920x1080p at MSAA 4X.

From http://www.gamespot.com/articles/witcher-3-dev-can-t-promise-improved-xbox-one-reso/1100-6426272/

Witcher 3 dev haven't tried DirectX12 and Witcher 3 XBO hasn't used Async compute.

YOU keep ignoring DirectX12 or any Async compute performance gains..

Using Async Compute on already low over head API (PS4) still increases 19 percent performance.

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#258  Edited By tormentos
Member since 2003 • 33793 Posts

@ronvalencia said:

Tell that NVIDIA's mobile Maxwellv1 GPUs.

The fact is 7770 is usually backed by stronger Intel CPU just like my laptop with Intel Core i7 3635QM CPU (2.4 Ghz with 3.4Ghz Turbo) and Radeon HD 8870M (7770 based GCN) graphics card.

My point with PC's minimum specs is about obtaining similar performance experience between PC hardware and counter your i3 arguments. Your i3 arguments are not applicable for DX12 Quantum Break.

There's "Screen Space" based shading LOL. Unless you have massive shading overdraw issues, you only shade the polygon surfaces that are visible. Deferred shading is a screen-space shading. Screen-Space Ambient Occlusion is another screen space shading, hence the part of it's name.

From http://www.gamespot.com/articles/witcher-3-dev-can-t-promise-improved-xbox-one-reso/1100-6426272/

Witcher 3 dev haven't tried DirectX12 and Witcher 3 XBO hasn't used Async compute.

YOU keep ignoring DirectX12 or any Async compute performance gains..

Using Async Compute on already low over head API (PS4) still increases 19 percent performance.

You know what the fun part will be when banchmarks for the game comes out and the i3 perform similar or better than those 6 cores AMD CPU.

I take CD project words over yours any day,so yeah having more triangles than you can shade defeat the whole purpose.

The xbox one already have those gains and you refuse to admit it,the xbox one is not PC full of pitfalls consoles have never have such horrible pitfalls and is the reason why 2 years + into the generation the xbox one keeps under performing hell in 2015 more games actually have both resolution and frame advantage over the xbox one than on 2014..lol

Don't worry as soon as quantum break benchmarks come out i get back to you.

Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#259  Edited By ronvalencia
Member since 2008 • 29612 Posts

@tormentos said:
@ronvalencia said:

Tell that NVIDIA's mobile Maxwellv1 GPUs.

The fact is 7770 is usually backed by stronger Intel CPU just like my laptop with Intel Core i7 3635QM CPU (2.4 Ghz with 3.4Ghz Turbo) and Radeon HD 8870M (7770 based GCN) graphics card.

My point with PC's minimum specs is about obtaining similar performance experience between PC hardware and counter your i3 arguments. Your i3 arguments are not applicable for DX12 Quantum Break.

There's "Screen Space" based shading LOL. Unless you have massive shading overdraw issues, you only shade the polygon surfaces that are visible. Deferred shading is a screen-space shading. Screen-Space Ambient Occlusion is another screen space shading, hence the part of it's name.

From http://www.gamespot.com/articles/witcher-3-dev-can-t-promise-improved-xbox-one-reso/1100-6426272/

Witcher 3 dev haven't tried DirectX12 and Witcher 3 XBO hasn't used Async compute.

YOU keep ignoring DirectX12 or any Async compute performance gains..

Using Async Compute on already low over head API (PS4) still increases 19 percent performance.

You know what the fun part will be when banchmarks for the game comes out and the i3 perform similar or better than those 6 cores AMD CPU.

I take CD project words over yours any day,so yeah having more triangles than you can shade defeat the whole purpose.

The xbox one already have those gains and you refuse to admit it,the xbox one is not PC full of pitfalls consoles have never have such horrible pitfalls and is the reason why 2 years + into the generation the xbox one keeps under performing hell in 2015 more games actually have both resolution and frame advantage over the xbox one than on 2014..lol

Don't worry as soon as quantum break benchmarks come out i get back to you.

To bad for you, somebody has done Async compute with a game and reduce frame time render.

YOU keep ignoring DirectX12 or any Async compute performance gains..

Using Async Compute on already low over head API (PS4) still increases 19 percent performance.

At March 2014 XDK, XBO's Async Compute features wasn't fully unlocked. XBO has it's first known Async compute usage with Rise of Tomb Raider XBO (November 2015) i.e. that's 21 months from initial feature unlocking to the release of Rise of Tomb Raider XBO.

There's only a small handful of PS4 games with Async compute. Note that Polaris GCN 4.0 includes improvements with it's sync graphics command processor.

From http://www.gamepur.com/news/16560-dev-explains-forza-horizon-2-1080p4x-msaa-management-says-controversial.html

It was also necessary to have a good understanding of what the bottlenecks were for each rendering stage so we could target ESRAM optimisations for systems that benefited from the additional bandwidth. ESRAM is pretty easy to manage, mainly because you don’t need to resolve textures to read them and you can make textures partially resident

For FH2, not just frame buffer for ESRAM i.e. it has PRT for it's ESRAM. PRT can be applied for DDRx or GDDR5 memory pools. The main point with PRT is to keep TMUs running near memory bandwidth limits and minimize stalls with texture loads from HDD/SSD storage.

I'm still waiting for PS4's 1920x1080p with MSAA 4X.

The main reason for XBO is under preforming PC's 7770 results are mostly due to the CPU issues. If XBO's CPU was powerful, they wouldn't need remote Intel Xeon servers for Crackdown 3.

If a developer wants to deliver improved game play logic/game play modeling over last gen consoles, the developer will put additional load on the CPU side.

For next gen 14 nm parts, AMD plans to replace Jaguar/Puma with low power ZEN variant. While AMD delivered improve CPU over IBM PPE/SPU, it's not enough.

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#260 tormentos
Member since 2003 • 33793 Posts

@ronvalencia said:

To bad for you, somebody has done Async compute with a game and reduce frame time render.

YOU keep ignoring DirectX12 or any Async compute performance gains..

Using Async Compute on already low over head API (PS4) still increases 19 percent performance.

At March 2014 XDK, XBO's Async Compute features wasn't fully unlocked. XBO has it's first known Async compute usage with Rise of Tomb Raider XBO (November 2015) i.e. that's 21 months from initial feature unlocking to the release of Rise of Tomb Raider XBO.

From http://www.gamepur.com/news/16560-dev-explains-forza-horizon-2-1080p4x-msaa-management-says-controversial.html

For FH2, not just frame buffer for ESRAM.

I'm still waiting for PS4's 1920x1080p with MSAA 4X.

The main reason for XBO is under preforming PC's 7770 results are mostly due to the CPU issues. If XBO's CPU was powerful, they wouldn't need remote Intel Xeon servers for Crackdown 3.

If a developer wants to deliver improved game play logic/game play modeling over last gen consoles, the developer will put additional load on the CPU side.

For next gen 14 nm parts, AMD plans to replace Jaguar/Puma with low power ZEN variant.

My god you repeat the same sh** over and over again even when you freaking know is wrong,FH2 has MSAAx 4 because it look like total sh** is a very unimpressive game,why doesn't F5 or F6 have MSAA X4 oh yeah they target 60FPS and better visuals so the xbox one can't do it.

I am still waiting for a next gen looking racer on xbox one 1080p and MSAA x4 did little for FH2 it look like total sh** compare to Drive Club.

No it under performs because it is weak as the 7770 is,in fact in flop count they are pretty much even but the 7770 has no ESRAM as pitfall,

Oh and physics can run on the GPU Sony has done it and they look quite great.

Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#261  Edited By ronvalencia
Member since 2008 • 29612 Posts

@tormentos said:
@ronvalencia said:

To bad for you, somebody has done Async compute with a game and reduce frame time render.

YOU keep ignoring DirectX12 or any Async compute performance gains..

Using Async Compute on already low over head API (PS4) still increases 19 percent performance.

At March 2014 XDK, XBO's Async Compute features wasn't fully unlocked. XBO has it's first known Async compute usage with Rise of Tomb Raider XBO (November 2015) i.e. that's 21 months from initial feature unlocking to the release of Rise of Tomb Raider XBO.

From http://www.gamepur.com/news/16560-dev-explains-forza-horizon-2-1080p4x-msaa-management-says-controversial.html

For FH2, not just frame buffer for ESRAM.

I'm still waiting for PS4's 1920x1080p with MSAA 4X.

The main reason for XBO is under preforming PC's 7770 results are mostly due to the CPU issues. If XBO's CPU was powerful, they wouldn't need remote Intel Xeon servers for Crackdown 3.

If a developer wants to deliver improved game play logic/game play modeling over last gen consoles, the developer will put additional load on the CPU side.

For next gen 14 nm parts, AMD plans to replace Jaguar/Puma with low power ZEN variant.

My god you repeat the same sh** over and over again even when you freaking know is wrong,FH2 has MSAAx 4 because it look like total sh** is a very unimpressive game,why doesn't F5 or F6 have MSAA X4 oh yeah they target 60FPS and better visuals so the xbox one can't do it.

I am still waiting for a next gen looking racer on xbox one 1080p and MSAA x4 did little for FH2 it look like total sh** compare to Drive Club.

No it under performs because it is weak as the 7770 is,in fact in flop count they are pretty much even but the 7770 has no ESRAM as pitfall,

Oh and physics can run on the GPU Sony has done it and they look quite great.

Why bring in 60 fps into this debate when PS4's The Order is also 30 fps? The shader power between the two boxes are about 30 percent difference i.e. XBO is 71 percent of PS4's shader power. XBO needs shaders cost to be 30 percent cheaper than PS4. For XBO's 1920x1080p frame buffer, shaders has to be 30 percent cheaper than PS4's version, then MSAA step is applied at the end of the rendering pipeline. The alternative is to reduce frame buffer resolution by 30 percent and run the same shader program version. My point, what's magical about another 30 percent extra shader power?

For The Order, PS4 has the CU shader/color-ROPS power to render 1920x1080p then apply MSAA but it's limited to 2X i.e. they blame the problem on memory bandwidth not CU power. If PS4's has 7950's memory bandwidth, MSAA 4X would be applied.

MSAA is a geometry edge AA that uses Z-ROPS(read depth data), color-ROPS units(read/write color data from/to frame buffer) and memory bandwidth. MSAA processors reads depth data, reads color data, apply MSAA and writes color data and it's memory bandwidth consumer. My point, MSAA power has nothing to do with CU power.

MSAA is applied at the end of the rendering pipeline. GCN's 32 ROPS was designed for MSAA with 7950/7970's level memory bandwidth and PS4's 32 ROPS doesn't have 7950/7970's memory bandwidth.

7770 doesn't have ESRAM pitfall hence no bandwidth mitigation when TMU and ROPS operates at the same time i.e. limited to ~72 GB/s of memory bandwidth for both TMU and ROPS.

The fact is 7770 is backed by stronger CPU.

At 1920x1080p, running my 7970 GE's memory bandwidth down to ~120 GB/s affects MSAA performance, but shader performance is largely unaffected.

I already know PS4 has Havok GPU and MSFT just taken over Havok. Hopefully, MSFT ports Havok GPU to PC world. Intel has a conflict of interest to not port Havok GPU to the PC.

Year 2006, before Intel

Loading Video...

Havok physics on the PC GPU back in 2006. Intel killed this development path on the PC.

Year 2006 is NVIDIA's Geforce 8800 GTX.

Havok Physics GPU revisited for PS4.

Loading Video...

Loading Video...

Sony already using MS Visual Studio for their PS4 development tool chain . Microsoft still sells Minecraft for PS4 and still publish by SCEE.

From http://blogs.microsoft.com/blog/2015/10/02/havok-to-join-microsoft/

As we welcome Havok to the Microsoft family, we will continue to work with developers to create great gaming experiences, and continue to license Havok’s development tools to partners. We believe that Havok is a fantastic addition to Microsoft’s existing tools and platform components for developers, including DirectX 12, Visual Studio and Microsoft Azure.

Havok is an amazing technology supplier in the games industry and the leading real-time physics creator. We saw an opportunity to acquire Havok to deliver great experiences for our fans. Throughout the company’s history, they’ve partnered with Activision, EA, Ubisoft, Nintendo, Sony, Microsoft and many others to create more than 600 games including Halo, Assassin’s Creed, Call of Duty, Destiny, Dark Souls and The Elder Scrolls.

From http://segmentnext.com/2015/12/21/power-of-the-cloud-is-available-to-ps4-if-sony-wants-it-no-xbox-app-for-mac-phil-spencer/

According to Microsoft, it is available to Sony or any other rival company for its games, even exclusives.

MS's power of the cloud (Azure service) is available to PS4, if Sony wants it.

Avatar image for tormentos
tormentos

33793

Forum Posts

0

Wiki Points

0

Followers

Reviews: 0

User Lists: 0

#262 tormentos
Member since 2003 • 33793 Posts

@ronvalencia said:

Why bring in 60 fps into this debate when PS4's The Order is also 30 fps? The shader power between the two boxes are about 30 percent difference i.e. XBO is 71 percent of PS4's shader power. XBO needs shaders cost to be 30 percent cheaper than PS4. For XBO's 1920x1080p frame buffer, shaders has to be 30 percent cheaper than PS4's version, then MSAA step is applied at the end of the rendering pipeline. The alternative is to reduce frame buffer resolution by 30 percent and run the same shader program version. My point, what's magical about another 30 percent extra shader power?

For The Order, PS4 has the CU shader/color-ROPS power to render 1920x1080p then apply MSAA but it's limited to 2X i.e. they blame the problem on memory bandwidth not CU power. If PS4's has 7950's memory bandwidth, MSAA 4X would be applied.

MSAA is a geometry edge AA that uses Z-ROPS(read depth data), color-ROPS units(read/write color data from/to frame buffer) and memory bandwidth. MSAA processors reads depth data, reads color data, apply MSAA and writes color data and it's memory bandwidth consumer. My point, MSAA power has nothing to do with CU power.

MSAA is applied at the end of the rendering pipeline. GCN's 32 ROPS was designed for MSAA with 7950/7970's level memory bandwidth and PS4's 32 ROPS doesn't have 7950/7970's memory bandwidth.

7770 doesn't have ESRAM pitfall hence no bandwidth mitigation when TMU and ROPS operates at the same time i.e. limited to ~72 GB/s of memory bandwidth for both TMU and ROPS.

The fact is 7770 is backed by stronger CPU.

At 1920x1080p, running my 7970 GE's memory bandwidth down to ~120 GB/s affects MSAA performance, but shader performance is largely unaffected.

I already know PS4 has Havok GPU and MSFT just taken over Havok. Hopefully, MSFT ports Havok GPU to PC world. Intel has a conflict of interest to not port Havok GPU to the PC.

Year 2006, before Intel

Havok physics on the PC GPU back in 2006. Intel killed this development path on the PC.

Year 2006 is NVIDIA's Geforce 8800 GTX.

Havok Physics GPU revisited for PS4.

Sony already using MS Visual Studio for their PS4 development tool chain . Microsoft still sells Minecraft for PS4 and still publish by SCEE.

From http://blogs.microsoft.com/blog/2015/10/02/havok-to-join-microsoft/

From http://segmentnext.com/2015/12/21/power-of-the-cloud-is-available-to-ps4-if-sony-wants-it-no-xbox-app-for-mac-phil-spencer/

According to Microsoft, it is available to Sony or any other rival company for its games, even exclusives.

MS's power of the cloud (Azure service) is available to PS4, if Sony wants it.

Yeah it shows is only 29% more powerful right.? Is that before or after you account for the xbox one pitfalls and cumbersome hardware.?

So how is 29% the difference yet in resolution the gap is 45% + from 2 to as much as 15FPS in some games.?

The Gap in performance in Project cars isn't 29%,that happen with many other games to like SWBF,COD and several others not only the PS4 has 45% resolution gap it also has from 2 to 8 FPS advantage which make that gap even bigger,so trying to say the xbox one is 71% of the PS4 shader power is just a sad way to try to downplay a gap that is showing more than 50% gap in performance.

Oh please should i queue UFC again and show you which has higher MSAA,? The xbox one has 2X the PS4 4X regardless of the PS4 rendering lower resolution which allow for 4x to be use the xbox one is 900p it SHOULD BE also 4x because Forza is 4x right and is 1080p...

There goes your argument again,but UFC doesn't look as shitty as Forza.

PS4 goes with a full 1080p output as expected, and treats it with a comprehensive pass of 4x MSAA - a multi-sample sweep that focuses on geometric aliasing.

http://www.eurogamer.net/articles/digitalfoundry-2015-final-fantasy-x-x2-hd-remaster-ps4-face-off

You claim you are waiting for a PS4 game with 1080p and 4xMSAA.? Oh yeah i already gave it to you some months back FF remasters is 1080p and 4xMSAA there goes your argument about bandwidth. You have some serious memory problems.

So yeah depending on what you do you can do 1080p and MSAA X4 which is why Forza Horizon 2 does it and the other forza games don't.

Does Ryse features MSAA x4.? Yeah i am sure it doesn't because the visuals already are taxing the unit to much regardless of bandwidth.

[UPDATE 13/2/15 17:00: Reviewing the assets, it looks like Ultra actually deploys 4x MSAA as opposed to a post-process solution, though some of the effects work appears to compromise the effect.]

http://www.eurogamer.net/articles/digitalfoundry-2015-vs-super-stardust-ultra-on-ps4

What another one...lol

Ill awaits your usual damage control and full page of irrelevant crap that has nothing to do with what is being argue.

:

Avatar image for ronvalencia
ronvalencia

29612

Forum Posts

0

Wiki Points

0

Followers

Reviews: 1

User Lists: 0

#263  Edited By ronvalencia
Member since 2008 • 29612 Posts

@tormentos said:
@ronvalencia said:

Why bring in 60 fps into this debate when PS4's The Order is also 30 fps? The shader power between the two boxes are about 30 percent difference i.e. XBO is 71 percent of PS4's shader power. XBO needs shaders cost to be 30 percent cheaper than PS4. For XBO's 1920x1080p frame buffer, shaders has to be 30 percent cheaper than PS4's version, then MSAA step is applied at the end of the rendering pipeline. The alternative is to reduce frame buffer resolution by 30 percent and run the same shader program version. My point, what's magical about another 30 percent extra shader power?

For The Order, PS4 has the CU shader/color-ROPS power to render 1920x1080p then apply MSAA but it's limited to 2X i.e. they blame the problem on memory bandwidth not CU power. If PS4's has 7950's memory bandwidth, MSAA 4X would be applied.

MSAA is a geometry edge AA that uses Z-ROPS(read depth data), color-ROPS units(read/write color data from/to frame buffer) and memory bandwidth. MSAA processors reads depth data, reads color data, apply MSAA and writes color data and it's memory bandwidth consumer. My point, MSAA power has nothing to do with CU power.

MSAA is applied at the end of the rendering pipeline. GCN's 32 ROPS was designed for MSAA with 7950/7970's level memory bandwidth and PS4's 32 ROPS doesn't have 7950/7970's memory bandwidth.

7770 doesn't have ESRAM pitfall hence no bandwidth mitigation when TMU and ROPS operates at the same time i.e. limited to ~72 GB/s of memory bandwidth for both TMU and ROPS.

The fact is 7770 is backed by stronger CPU.

At 1920x1080p, running my 7970 GE's memory bandwidth down to ~120 GB/s affects MSAA performance, but shader performance is largely unaffected.

I already know PS4 has Havok GPU and MSFT just taken over Havok. Hopefully, MSFT ports Havok GPU to PC world. Intel has a conflict of interest to not port Havok GPU to the PC.

Year 2006, before Intel

Havok physics on the PC GPU back in 2006. Intel killed this development path on the PC.

Year 2006 is NVIDIA's Geforce 8800 GTX.

Havok Physics GPU revisited for PS4.

Sony already using MS Visual Studio for their PS4 development tool chain . Microsoft still sells Minecraft for PS4 and still publish by SCEE.

From http://blogs.microsoft.com/blog/2015/10/02/havok-to-join-microsoft/

From http://segmentnext.com/2015/12/21/power-of-the-cloud-is-available-to-ps4-if-sony-wants-it-no-xbox-app-for-mac-phil-spencer/

According to Microsoft, it is available to Sony or any other rival company for its games, even exclusives.

MS's power of the cloud (Azure service) is available to PS4, if Sony wants it.

Yeah it shows is only 29% more powerful right.? Is that before or after you account for the xbox one pitfalls and cumbersome hardware.?

So how is 29% the difference yet in resolution the gap is 45% + from 2 to as much as 15FPS in some games.?

The Gap in performance in Project cars isn't 29%,that happen with many other games to like SWBF,COD and several others not only the PS4 has 45% resolution gap it also has from 2 to 8 FPS advantage which make that gap even bigger,so trying to say the xbox one is 71% of the PS4 shader power is just a sad way to try to downplay a gap that is showing more than 50% gap in performance.

Oh please should i queue UFC again and show you which has higher MSAA,? The xbox one has 2X the PS4 4X regardless of the PS4 rendering lower resolution which allow for 4x to be use the xbox one is 900p it SHOULD BE also 4x because Forza is 4x right and is 1080p...

There goes your argument again,but UFC doesn't look as shitty as Forza.

PS4 goes with a full 1080p output as expected, and treats it with a comprehensive pass of 4x MSAA - a multi-sample sweep that focuses on geometric aliasing.

http://www.eurogamer.net/articles/digitalfoundry-2015-final-fantasy-x-x2-hd-remaster-ps4-face-off

You claim you are waiting for a PS4 game with 1080p and 4xMSAA.? Oh yeah i already gave it to you some months back FF remasters is 1080p and 4xMSAA there goes your argument about bandwidth. You have some serious memory problems.

So yeah depending on what you do you can do 1080p and MSAA X4 which is why Forza Horizon 2 does it and the other forza games don't.

Does Ryse features MSAA x4.? Yeah i am sure it doesn't because the visuals already are taxing the unit to much regardless of bandwidth.

[UPDATE 13/2/15 17:00: Reviewing the assets, it looks like Ultra actually deploys 4x MSAA as opposed to a post-process solution, though some of the effects work appears to compromise the effect.]

http://www.eurogamer.net/articles/digitalfoundry-2015-vs-super-stardust-ultra-on-ps4

What another one...lol

Ill awaits your usual damage control and full page of irrelevant crap that has nothing to do with what is being argue.

:

Recent games shows about 30 percent difference i.e. 1920x1080p vs 1600x900p. You know XBO (and PS4) has weaker CPU than gaming PCs powered by Intel Sandybridge (with AVXv1)and XBO (and PS4) games needs special CPU multithreading handling i.e. they may have X86-64 with AVXv1 CPUs but both consoles still carries multithreading needs from last gen X360/PS3 consoles.

Final Fantasy X/X2 was initially remastered ports from PS2 to PS3 and PS4 still shows low polygon (affects depth data) and textures (affects TMUs) from PS3. Final Fantasy X/X2 made zero effort to use tessellation (geometry amplification).

To back this, textures are digitally enhanced to create a more vibrant look across the board. The base resolution of almost every texture map is identical to the PS3's, but each is treated to achieve a higher-contrast look - also reinforced by a superior grade of anisotropic filtering.

Superior grade anisotropic filtering (AF) is like when a PC's higher grade GPU runs legacy games with a driver settings override. They added ambient occlusion i.e. a known shader effects.

The art assets storage foot print for Forza Horizon 2 is 35.84 GB which is not comparable to Final Fantasy X/X2 Remastered's 22.9 GB and it's filled with plenty FMVs. A native DX11 title uses higher grade texture compression than DX9 version.

For example, Final Fantasy Lightning Returns PC version (ported from Xbox 360 or PS3) has recommended specs of Radeon HD 5770 (renamed to 6770) GPU with art assets storage foot print 20.9 GB and it's filled with plenty FMVs (5 GB of videos) which is similar to Final Fantasy X/X2 Remastered. The actual game is about 15 GB. Final Fantasy Lightning Returns PC edition uses DX9c.

Loading Video...

Are you nuts? The render is like a PC playing a legacy game and it's worst than last gen Final Fantasy Lightning Returns port to PC.

Based from Final Fantasy Lightning Returns PC version, Final Fantasy X/X2 Mastered edition PS4 memory bandwidth load is not comparable to Forza Horizon 2 which used both ESRAM (render targets and PRT) and 256 bit DDR3-2133 memory bandwidth.

The Order's choice is either 1920x1080p MSAA 2X or 1920x800p MSAA 4X and the wall was memory bandwidth. The Order used Forward Plus renderer as per AMD's advice.

My main point, different machine strengths affects different rendering times in each rendering stages e.g.

Gaming PC with Sandybridge i5/i7 will resolved CPU side geometry dot products faster than console's CPU.

Gaming PC with Sandybridge i5/i7 will resolved CPU side geometry dot culling faster than console's CPU.

Gaming PC with Sandybridge i5/i7 will resolved CPU side soft-render approximation faster than console's CPU. Good 3D engine works out what's to feed the GPU i.e. minimise overdraws. Gaming PC's weakness with API overheads was removed with DirectX12.

A GPU with higher shader power resolves shader operations faster than GPUs with lesser shader power. Developers already stated PS4 has this advantage over XBO. PC GPU with 7950 or higher beats PS4.

A GPU with higher memory bandwidth and internal bus completes read and write operations faster than GPUs with lesser memory bandwidth and internal bus. Developers already stated XBO has this advantage over PS4 but with caveats. If the GPU is bottlenecked by shader power, memory writes will be gimped. PC GPU with 7950 or higher beats XBO in this area. For FH2, the programmers made sure shader budget wouldn't gimp XBO's memory bandwidth writes.

Why 7950? 7950 GPU is the faster solution in most XBO and PS4 strengths. GPU such as 7870 XT or 7870 has stronger shader power. Geometry 3D engine in PC GCN 1.0/1.1(two triangles per cycle model), XBO and PS4 are the same. ROPS and Geometry 3D engine separates the GPUs from DSPs (e.g. IBM SPE).

For building a gaming box with smaller budget, I would select shader power over high memory bandwidth i.e. I partially agree with Sony's design choices for PS4 and I don't fully agree with AMD's HSA PR.

Avatar image for FireEmblem_Man
FireEmblem_Man

20385

Forum Posts

0

Wiki Points

0

Followers

Reviews: 7

User Lists: 0

#264 FireEmblem_Man
Member since 2004 • 20385 Posts

LOL @tormentos Getting owned be a hardware genius! Give it up Tomato, you don't know jack sh!t about hardware at all