You need to post whatever you have because you are interpreting it wrong.
-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
EuroGamer: More details on the BALANCE of XB1
- Thread starter artist
- Start date
You need to post whatever you have because you are interpreting it wrong.
NTIETS13A_Superiority
Banned
MS could not use those 2 and this is upclock was better for them is bs (their games may have had gains from upclock but thats beside the point) CELL had terrrible yields and it used 1 out of 8 for better yields MS uses 1 out of 7. Who in their right mind would want 1/7 of CU silicon go "waste"?
gundamkyoukai
Member
Which just goes back to what Cerny said that if you have extra cycles you can use them for GPGPU.
I would hope that people would know your never maxing out the GPU all the time.
PS4 unbalanced confirmed
that's really interesting though. And perhaps would that mean that cross-gen multiplatform titles might not show much difference between xbox and PS4? Assuming they are relatively straightforward up-ports.
Actually, if you are not CU bound on - say - a 'typical' 1080p game today on either system, but are fillrate bound, it would spell bigger trouble for XB1.
The implications of MS's comments aren't really flattering on a number of fronts.
If I was MS, I'd be hoping PS4 was CU bound too...else, that much larger ROP advantage could be brought to bear in a more straightforward fashion.
It's also funny to see them say 'look, Sony thinks we're right too!', when Sony's prescription for balance in the typical case for rendering actually looks a whole lot different than MS's. The funny thing about MS talking about balance while looking at one part of the pipeline...
(In reality, personally, I think the reality of performance is very much a ymmv thing...at a given res some games will be more CU bound, some games will be less ROP bound etc. etc. In any case, all these variations will do better on PS4)
He was just making clear that performance doesn't scale linearly, which is true. He was saying that fact might encourage devs to use that performance for compute instead of rendering. You aren't forced into anything. I don't think devs would be reporting 40-50% gains over XBO performance if 14 was the max they could use efficiently.
Also this:
http://www.scei.co.jp/corporate/release/130221a_e.html
so to paraphrase
"you could use those extra CUs for graphics, but you'll hit a bit of diminishing returns - wouldn't that be a waste? Sure your game will look a little better, but imagine how amazing it'll look if you explored using those CUs for GPGPU instead. All that amazing physics, AI, fluid dynamics - it'll look fantastic"
"you could use those extra CUs for graphics, but you'll hit a bit of diminishing returns - wouldn't that be a waste? Sure your game will look a little better, but imagine how amazing it'll look if you explored using those CUs for GPGPU instead. All that amazing physics, AI, fluid dynamics - it'll look fantastic"
That's how I understand it, yes.
SenjutsuSage
Banned
This isn't my interpretation. It's the interpretation of an experienced games programmer that learned this at an official Sony devcon. I'm just repeating their understanding, so there's very little getting lost in translation as a result. Only way the info could be misinterpreted is if they misinterpreted it, which I doubt.
More or less, absolutely.
Well then they are misinterpreting it or you are not getting the full story because half of what you post is completely false.
shinra-bansho
Member
Is the developer conference in question GDC and is the presentation in question "Overview of the PS4 for Developers" because then I could understand why it couldn't be shared - since it's behind a paywall on their GDC vault.
From a review of the 7790, framerates (not the ideal comparison of course, since the 7850 has other advantages beyond the CU count, but the 7790 has the magic number of CUs)
What's seemingly being described here seems beyond just diminishing returns - it's being made out as if improvement falls entirely off a cliff after 14 CUs - like it's the magic number, while AMD is happily churning out 16, 20, 32 CU cards etc.
From a review of the 7790, framerates (not the ideal comparison of course, since the 7850 has other advantages beyond the CU count, but the 7790 has the magic number of CUs)
Code:
Radeon HD 7790 1GB Radeon HD 7850 1GB
BioShock Infinite 60.5 81.5
Hitman Absolution 35.5 46.7
Batman: Arkham City 65 89
Tomb Raider 53.4 69.4
Sleeping Dogs 45.9 56.3
Metro 2033 23.7 32.6Sure. That's all well and good and seems plausible. But that's not really the only claim being made here.
As above, seems completely plausible - if the extra compute resources aren't really going to do much for a given piece of software in terms of graphics it makes perfect sense to utilise them for something else otherwise it's just wasted resources. Of course, as also noted above, that doesn't seem to be the sole claim being made here - which seems to be more along the lines of the PS4 is somehow different from desktop GPUs and beyond using 14 of its CUs for graphics the performance increase plummets to the realms of negligibility.
Also very true. If a multiplatform game really is utilising 400 GFLOPS of the GPU for non-graphical tasks, it probably doesn't end well for the XB1.
What are you talking about?
That's the conclusion that you came to in your head I never said anything like that, I just pointed out that even if you split the GPU up into 14 + 4 it would still be 400GFLOPS of compute left over for the devs to use.
I know what a GPGPU is.
could Next Gen see 1 of the biggest improvements between console gens do to GPGPUs?
GPGPU Computing & why you should be more excited about Kinect 2 & the Next PS-Eye.,
If that were true, it would be pretty damned troublesome. Happily, it's bullshit.
Doesn't this sound eerily similar to Penello's "I've got a technical fellow" argument? I'm sorry Senjutsu, but that justification holds no water here. When information changes hands, our own presumptions invariably crop in. Going by your last few posts, you are paraphrasing, not reporting. So there is always a possibility of misinterpretation.
NTIETS13A_Superiority
Banned
If bandwith allows, use them for graphics if graphics is what you want. With similar engines with XO and PS4 mp games I would assume diminishing returns would come at some point. Depends on scalability of tasks and if extra work gives noticeable results. I can not wait to desing my own engine.
A game's sweet spot for rendering actually wouldn't have to go far north of 14 to achieve that kind of difference, if rumours about XB1's OS reservations are true.
Say - 15 800Mhz CUs vs 10.8 853 Mhz CUs (12-10%) with 32 ROPs vs 16 ROPs...that would give a pretty straight forward 30-90% gain on the shading and pixel output end of the pipeline, where perf might land depending on where in that segment of the pipeline the bound balanced out.
And this example is assuming a case where rendering perf doesn't scale at all past a certain amount of ALU, whereas realistically we might be talking more about a less than linear scaling (i.e. some further improvement on the lower end of that range).
So back of a napkin stuff - but I don't think this idea is necessarily inconsistent with some of the rumours we've been hearing about rendering deltas. Remember too: it all depends on the shape of an individual game's pipeline.
Cool, you know where to find me then.
SenjutsuSage
Banned
Just sent.
Keep in mind I didn't say it couldn't be used however a dev wished. Just what Sony presented to devs about a bit of diminishing returns for graphics operations after the 14 CU mark. Devs if they wish can use all 18 for graphics, but apparently there's a significant drop off in the benefit that you see for the additional ALU resources for graphics past the 14 CU mark, which is why Sony suggests the most optimal use for the extra 4 CUs is to use them for compute work.
CadetMahoney
Member
They keep trying to defend it somehow meaning they're scared it will put off potential buyers, since they know ppl don't want to spend $60 for a game and get an inferior product. Let alone $500 up front for an inferior product. Leaving a hugely dissatisfied customer. So they're trying to muddy the waters as much as possible hoping xbone buyers will be happy with their product.
They're desperate and clasping at straws.
Finally...one way or other this will be settled.
shinra-bansho
Member
What exactly is being sent if what we're currently discussing is just a verbatim repetition of an interpretation of a developer presentation and not any form of official document? Like an ordinary PM with the same information that's been posted in the thread? :/
Oh well, iirc, bish is a developer as well...? So it would be interesting to get his take on this all anyway.
Oh well, iirc, bish is a developer as well...? So it would be interesting to get his take on this all anyway.
I'm glad you learned your lesson SenjutsuSage. Spinning makes your head dizzy.
panda-zebra
Banned
Pirelli manufacture the last 4 CUs CONFIRMED.
Basically!
& it's also the vision I seen
could Next Gen see 1 of the biggest improvements between console gens do to GPGPUs?
NTIETS13A_Superiority
Banned
Yeah he claims to know something and is willing to only give his word, no links... I hope mods do not go too easy on that shit.
The whole notion of discrete distribution of CUs in this conversations is stupid though. Async compute is there so you don't have to do that.gofreak said:
Eg. in a deferred renderer all CUs will be effectively idle during Z-prepass(meaning ~18 would be available for compute during), and all of them could be utilized during shading pass.
Sure it may average out to 4:14 over the course of frame - but if you discretely partition it, chances are you're losing performance on both sides.
I'm not sure that claim is being made. The same kind of behaviour about non linear return on ALU in 'typical games' can be observed on PC, in that DF comparison article for example.
Sure, but it's just easier to talk about in terms of discrete CUs to try and quantify the amount of performance that could be going a-begging. But yeah, that is worth clarifying. No one's going to actually be partitioning CUs, I don't think that's even possible through the API (?)
I believe he is a developer for a well known studio but for obvious reasons doesn't say for which one .
So it would be easy for him to check if someone is making up stuff or not through his connections.
Feel free to correct me if i got this wrong . Bish.
LittleJohnny
Banned
And IIRC, these 4 extra CU are not the same as the other 14, they're specialized for compute which would make it an even bigger waste.
Which makes no sense and no one can explain how.
All 18 CUs are the same.
gundamkyoukai
Member
There not specialized for anything , i can't believe we back to this .
This has been debunk already and explained .
EDIT Fafalada just gave us a perfect eg of how things can go .
Not 100% round = 100% round?
NTIETS13A_Superiority
Banned
Wouldn't this affect any number of CUs? So it would be so with XO too? That would not be what we are thinkering here. And wouldnt constant, say 4 CUs for compute better as you have more reliable access to resources? Are engines more foward or deferred now days? How would f-rendered engines fare on XO and PS4?
shadowsdarknes
Member
I'm seeing Microsoft fans on other boards claiming this article is stating that the X1 is officially proven more powerful. lol
But that doesn't make sense, that 14 is some kind of magic number. It will completely depend on your game engine. That's like someone saying "after 6 GB of RAM you have diminishing returns in a PC". Well, sure, maybe if you're only using it to read your email, but if you're running a DB on that machine you'l easily soak up the 6GB.
And you didn't even claim diminishing returns, originally, you said this:
NTIETS13A_Superiority
Banned
Well you remember wrong and this misinformation came from vgleaks.. cerny debunked this ages ago.
SenjutsuSage
Banned
I did
That explains exactly diminishing returns. There is a drop off in the value of the extra ALU resources for graphics after a certain point, that point being 14 CUs.
There are no docs. He will probably explain what he heard and reveal his source to Bish in the process.
SwiftDeath
Member
Hopefully bish gives us his interpretation of it as I think there is some truth to what Senjutsu said but likely not quite as big a deal as originally suggested
Curious to know what the returns are like from those 4 CU's
Although not sure I'd understand it to be honest
Nothing about diminishing returns was ever stated by anyone but ms pr guys. the changes made by sony would not be needed if you were always have some cu just for gpgpu work. Not buying any of this nonsense.
Yes but distinction is important when arguing about "wasted CU potential in graphics", as scenarios where you target max consumption at different parts of frame are quite viable in closed boxes, even if bandwith/fill suggests you can't in average case.gofreak said:
No. Games are inherently spiky when it comes to performance, nothing is better off with evenly distributed average, that's the entire reason why we have unified shaders since 2005.NTIETS13A_Superiority said:
But yes, XB1 will operate like that as well (barring the differences in approach to ACU by each company).
I did
That explains exactly diminishing returns. There is a drop off in the value of the extra ALU resources for graphics after a certain point, that point being 14 CUs.
Tbh my interpretation (fwiw) was that 14 was already past a very sharp drop off...now you're sounding like "a bit of a drop past 14" although I confess I have no idea where 'right of the knee' is
NTIETS13A_Superiority
Banned
And that is complete BS. It depends on application/tasks. I forgot we have ignorelist.
Well it's a bit harsh to call bs until bish confirms either way isn't it...
They are all the same.
basically 14 CU's is good enough for 1080P @ 60FPS using it for just fixed function graphics, so the other 4 can get you something like 1080P @ 75FPS or 1400P @ 60FPS.
but why would you do that if they are not standards? so you will be better off using the 4 CU's to add more effects using compute.
The DF comparison that tried to isolate 50%+ ALU showed a range of returns in 5 of today's games at some different levels of settings and effects. Those games showed a circa 17-30% improvement with 50% more ALU.
i.e. non-linear, 'might be better of spending it on something else' in some cases, but still fairly substantial
SEGAvangelist
Member
You'll still see him quoted and when you don't you'll wonder what people are responding to... ignore won't work.
Similar threads
- Poll