Ricerocket
Member
Nice misterxmedia picture.
Glad it's clear we know your source of news.
I don't understand any of this hardware talk. Can someone tell me (theoretically) what sort of differences we will notice between xbone and ps4 multiplats? How noticeable do yu think they'll be? I know this is probably a difficult question to answer but I'm really torn on which preorder to keep ( impending baby means I can only justify one!), my head says ps4 but I have friends who live a few hundred miles away going for the xbox and I'll miss not being able to play with them. On the flip side I'll feel gypped that I'll be playing a gimped version of the game. Decisions decisions...![]()
That's not my point.
I'm not interested about how many CUs have been around.
I'm interested to understand how sonyboys see Cerny debunked only the 14+4 among all the +9877863248732642768e246% leaks.
Did he just link us to misterxmedia? Holy shit!
Lmao at the 14+4 CU claims although it's already been disputed a million times.
What I find funny is even if the 14+4 CU split was true that's still 18 CU vs 12 CU
14+4 CU vs 12CU
That some how make things balance?
WTF am I reading?
http://ic.pics.livejournal.com/misterxmedia/21549619/79881/79881_original.jpg
That's not my point.
I'm not interested about how many CUs have been around.
I'm interested to understand how sonyboys see Cerny debunked only the 14+4 among all the +9877863248732642768e246% leaks.
In what way do they look technically superior?
In his eyes
the light, the heat
in his eyes, it is complete
that is an image posted on the misterX blog attempting to blow smoke so that we dont focus on how his house of cards regarding the Xbone hardware falls apart around him
Lol I can just imagine your face!Nobody really knows until the games come out, on paper PS4 has the raw power advantage but that's never a guarantee that it will show to any huge degree in multiplatform games.
And I wouldn't worry, I've spent the last seven years playing the 'gimped' versions of the big releases and apart from being utterly disgusted by the disastrous Bayonetta port (and some enemy-heavy parts in the Orange Box when the frame rate went further south than my gran's tits) I haven't had such a terrible time. Unless you put both version side-by-side you can adjust to a shittier frame rate, or fewer shrubs in most games.
I can never forgive my PS3 for that Ghostbusters rug though. It was worse than the time I played Ghostbusters on my mate's C=64 and loved the intro so much I asked for the Spectrum version for Christmas.
I mean listen:
http://www.youtube.com/watch?v=l-0-Us_jumQ
http://www.youtube.com/watch?v=Hz5VvMItHmM
/gutted kid
Did he just link us to misterxmedia? Holy shit!
![]()
That's not my point.
I'm not interested about how many CUs have been around.
I'm interested to understand how sonyboys see Cerny debunked only the 14+4 among all the +9877863248732642768e246% leaks.
You can't be serious
With any other GAFfer I'd assume it to be a joke, but with his post history...
I just got brain cancer trying to read his blog, he thinks that the GPU is now based off R9 280x which has the "beefed up CUs" of the R9 290x and all the chips added together gives the Xbone 4.5-5TF of power.
My point is, I've never known a point in history where it's not been made quite clear which console is more technically capable, right from the very onset. Why is it different now?
The suggestion is that developers aren't pushing the machines, (which applies for both) but surely Sony want the graphics to impress? They must be aware that graphics sell machines to the average gamer?
There's developing titles and there's developing titles, are we talking AAA first party/exclusive titles here which are effectively selling the next-generation to us? because that is really what we are discussing.
Seagate Laptop Thin SSHD 1TB
SATA 6Gb/s with NCQ for interface speed, 64MB, 5400RPM, 2.5"
Cheap too.
Check his post again - he was looking for 64GB of cache. Something big enough for a BD's worth of a game to sit in. 64MB is nothin'.
And now we reached a new low level in discussion here... sad.
That's not my point.
I'm not interested about how many CUs have been around.
I'm interested to understand how sonyboys see Cerny debunked only the 14+4 among all the +9877863248732642768e246% leaks.
We understand that Sony found that using 4 CUs in GPGPU tasks can be more efficient and with better overall results than using all of them for rendering tasks, but they dont stop developers to use them only for rendering or GPGPU, as they wish depending on the game demands. Its not mandatory at all. Sony is only providing more resources to the developers.
If you want to debate the merits of using all 18 CUs for rendering vs. dedicating some (4) for GPGPU, that's another discussion.
vgleaks said:We understand that Sony found that using 4 CUs in GPGPU tasks can be more efficient and with better overall results than using all of them for rendering tasks,
That's not my point.
I'm not interested about how many CUs have been around.
I'm interested to understand how sonyboys see Cerny debunked only the 14+4 among all the +9877863248732642768e246% leaks.
Read this post:That's not my point.
I'm not interested about how many CUs have been around.
I'm interested to understand how sonyboys see Cerny debunked only the 14+4 among all the +9877863248732642768e246% leaks.
What that slide is saying (and yeah, it's old info, so who knows by now) is that the rest of the GPU hardware is optimally matching the typical rendering throughput as if there were 14CUs. So if you use 18CUs just for rendering, the takaway is that that you'd run into some other bottleneck which would prevent the amount of the speed boost you'd expect from those CUs. So the idea is that the developers would be encouraged to use those 410GF for compute instead - for stuff like what the resogun is doing for instance, raytraced reflections or whatever else.
What that slide is not saying is that the mentioned 4CUs have anything about them that's different in terms of hardware or implementation. You can use them all for rendering, but the idea is that the few of them would be better utilized if they're used for compute.
I wish we'd hear some developers experience with real life hardware on this front.
It doesn't mean that. Every CU is the same.
All that's being said is that for a typical game targeting a given res (let's say 1080p) - however they're defining typical - there's more CU time available than you might strictly need for just graphics. Or to put it another way, there's possibly CU time going to waste in 'a typical game'. That past a certain point you're probably going to be bound by some other part of the pipeline anyway, and throwing more CU at the problem past that point will see less than linear returns. And so you may as well throw CU time at other tasks in that case.
But as Cerny said, it's not a formalisation in hardware of any split or any target for devs. If a game's graphics pipeline was compute heavy, the graphics pipeline alone could scale well beyond that suggested amount of CU time. For devs, their mileage may vary - it's extremely software dependent. However Sony is trying to encourage devs to throw in compute tasks because it could be a disproportionate competitive strength for PS4.
At this point, you don't even have a worthwhile argument. You're just yelling at the wind over nothing in particular.
Look, I'll help you out a bit - VGLeaks themselves put up a handy article concerning the source of the 14+4 number.
If you want to debate the merits of using all 18 CUs for rendering vs. dedicating some (4) for GPGPU, that's another discussion. Don't try to peddle the 14+4 hard split myth.
Actually haha, I expected that reaction to be fair. For the record I don't believe anything on his site, I wouldn't of used that picture had it not had the tweets from the IGN editor which was the point of the post. If I could of found them separately then I would of used them but that was just easier.
No
Cerny never debunked anything. He said about the rumor, but no numbers were given about PS4 GPU by Sony, except the gflops.
And also that subjects could change without previous information.
All those +98893743897432% specs were given by VGLeaks.
There they go.2dshmuplover
Banned
(Today, 03:35 PM)
Actually haha, I expected that reaction to be fair. For the record I don't believe anything on his site, I wouldn't of used that picture had it not had the tweets from the IGN editor which was the point of the post. If I could of found them separately then I would of used them but that was just easier.
Actually haha, I expected that reaction to be fair. For the record I don't believe anything on his site, I wouldn't of used that picture had it not had the tweets from the IGN editor which was the point of the post. If I could of found them separately then I would of used them but that was just easier.
Surely this is horseshit from VGLeaks? How can Sony admit to it being more efficient to augment the rest of the system by sacrificing a few CU's for other tasks, it flies in the face of the gist of the MS rebuttle in this thread.
Am confused now..
Now when I say that many people say, "but we want the best possible graphics". It turns out that they're not incompatible. If you look at how the GPU and its various sub-components are utilised throughout the frame, there are many portions throughout the frame - for example during the rendering of opaque shadowmaps - that the bulk of the GPU is unused. And so if you're doing compute for collision detection, physics or ray-casting for audio during those times you're not really affecting the graphics. You're utilising portions of the GPU that at that instant are otherwise under-utilised. And if you look through the frame you can see that depending on what phase it is, what portion is really available to use for compute.
PS4 is capable of a feature that Cerny calls "asynchronous fine-grain compute". On older GPUs you have a command processor that has a graphics pipeline for rendering and a compute pipeline for GPGPU. You can only do either rendering or GPGPU efficiently, both at the same time is pretty ineffecient.
AMD's GCN GPUs have asynchronous compute engines (ACEs). Basically, these ACEs are like additional compute pipelines for GPGPU. You can use them instead of the compute pipeline in the command processor. As a result, the latter will be able to focus on maximum rendering performance. A HD7970 has 2 of these ACEs, it can take care of 2 compute jobs at the same time. PS4 has 8 ACEs and can take care of 8 compute jobs at the same time.
To minimize the overhead for graphics rendering, you want to split your compute load into as many small jobs as possible (fine-grain) and "feed" it to your shader cores when they're not fully utilized during rendering jobs (which happens all the time on modern GPUs). Therefore each ACE has compute queues. A HD7970 only has 2 queues (one queue for each ACE), a PS4 has 64 compute queues (8 queues for each ACE). That means programmers can queue up to 64 different compute jobs at the same time in PS4. The 8 ACEs will choose the right jobs at the right time based on pre-defined dependences and according to Cerny you will not notice a penalty for graphics rendering since the ACEs wait until some part of the GPU is underutilized.
Mark Cerny also said, that he expects studios to use this feature to the full extent in two or three years.
Actually haha, I expected that reaction to be fair. For the record I don't believe anything on his site, I wouldn't of used that picture had it not had the tweets from the IGN editor which was the point of the post. If I could of found them separately then I would of used them but that was just easier.