By now enough time has passed that I can come back here and hopefully answer/not answer a few questions :)
In the 8+ years I've been running this place, I don't think I've ever pulled an article before. I can't be too specific here, but some folks needed to be kept anonymous and I had to make a decision for the greater good in the long run. I apologize for remaining quiet about it for so long, but it was necessary.
With that out of the way - there's a lot to talk about.
I finally managed to pry a pair of 7800 GTXs away from Derek's hands and I've been working to answer the question of how fast of a CPU need to feed these things. There are a number of variables that have to be taken into account, most importantly being the resolution you're running at. The thing that's truly new about a card as powerful as the G70 is that you really start being limited by your monitor in what resolutions it supports. While owners of large analog CRTs have a lot of flexibility with what resolutions they can run at, LCD owners can't; so if you've got a G70 hooked up to a 1600 x 1200 panel you'll have to make different CPU decisions than if you have a 1920 x 1200 panel. I'm trying to simplify the decision making as best as possible and for this round I'm only focusing on single card solutions, but if there's demand later I can tackle SLI requirements.
I finally hooked up the G70 to the 30" Cinema Display and gave Doom 3 a whirl at 2560 x 1600. What I find most interesting is that once you start getting far above 1600 x 1200 it's no longer about making the game look good, it's about making the game look good on your monitor. For example, there's not too much difference playing Doom 3 at 1920 x 1200 vs. 2560 x 1600, it's just that the former looks great on a 24" monitor while the latter looks great on a 30" monitor. The quest for perfect image quality stops being about resolution and starts being about screen size; almost in a way similar to how consoles used to be, where your only hope for a "better" picture was to go to a larger screen, since you couldn't control resolution.
The pendulum will swing away from ultra high resolutions as games become more and more demanding. There are still some titles that even the G70 can't handle at above 1280 x 1024.
Monday's Athlon 64 Memory Divider article has got me thinking a lot about multitasking and its impacts on higher speed memory. Theoretically there should be some pretty big differences between DDR400 and DDR500 once we get into the heftier multitasking scenarios, but I want to get an idea of exactly how widespread that need is. My initial tests only revealed one scenario where there was a tangible performance boost, but I think they warrant some additional testing. After I'm done with this memory divider stuff I'll head on to that.
Many of you have asked for a Battlefield 2 CPU scaling article and I'm more than happy to oblige, so I've started working on the planning for such an article. Right now I'm stuck trying to figure out how best to make it a manageable benchmarking task, as I'd like to be able to provide accurate CPU/GPU recommendations for each performance class. I think I'll inevitably have to limit what GPUs I cover, but I'll do my best to include the ones you guys want the most.
I've been stuck on a H.264 kick for a while now, so I figured that doing a CPU comparison involving H.264 would be something interesting to do. My only question, other than Quicktime 7 and Nero, what are you folks using to encode H.264 on the PC?
Remember Gigabyte's i-RAM from Computex? Well, one should be in my hands very soon and given the interest in it, it's going to receive top priority as soon as I've got it. Which begs the question, are there any particular tests you all would like to see? I'll admit, I am a bit surprised by the positive response the i-RAM received; I expected people to be interested in it, just not this interested in it.
In the 8+ years I've been running this place, I don't think I've ever pulled an article before. I can't be too specific here, but some folks needed to be kept anonymous and I had to make a decision for the greater good in the long run. I apologize for remaining quiet about it for so long, but it was necessary.
With that out of the way - there's a lot to talk about.
I finally managed to pry a pair of 7800 GTXs away from Derek's hands and I've been working to answer the question of how fast of a CPU need to feed these things. There are a number of variables that have to be taken into account, most importantly being the resolution you're running at. The thing that's truly new about a card as powerful as the G70 is that you really start being limited by your monitor in what resolutions it supports. While owners of large analog CRTs have a lot of flexibility with what resolutions they can run at, LCD owners can't; so if you've got a G70 hooked up to a 1600 x 1200 panel you'll have to make different CPU decisions than if you have a 1920 x 1200 panel. I'm trying to simplify the decision making as best as possible and for this round I'm only focusing on single card solutions, but if there's demand later I can tackle SLI requirements.
I finally hooked up the G70 to the 30" Cinema Display and gave Doom 3 a whirl at 2560 x 1600. What I find most interesting is that once you start getting far above 1600 x 1200 it's no longer about making the game look good, it's about making the game look good on your monitor. For example, there's not too much difference playing Doom 3 at 1920 x 1200 vs. 2560 x 1600, it's just that the former looks great on a 24" monitor while the latter looks great on a 30" monitor. The quest for perfect image quality stops being about resolution and starts being about screen size; almost in a way similar to how consoles used to be, where your only hope for a "better" picture was to go to a larger screen, since you couldn't control resolution.
The pendulum will swing away from ultra high resolutions as games become more and more demanding. There are still some titles that even the G70 can't handle at above 1280 x 1024.
Monday's Athlon 64 Memory Divider article has got me thinking a lot about multitasking and its impacts on higher speed memory. Theoretically there should be some pretty big differences between DDR400 and DDR500 once we get into the heftier multitasking scenarios, but I want to get an idea of exactly how widespread that need is. My initial tests only revealed one scenario where there was a tangible performance boost, but I think they warrant some additional testing. After I'm done with this memory divider stuff I'll head on to that.
Many of you have asked for a Battlefield 2 CPU scaling article and I'm more than happy to oblige, so I've started working on the planning for such an article. Right now I'm stuck trying to figure out how best to make it a manageable benchmarking task, as I'd like to be able to provide accurate CPU/GPU recommendations for each performance class. I think I'll inevitably have to limit what GPUs I cover, but I'll do my best to include the ones you guys want the most.
I've been stuck on a H.264 kick for a while now, so I figured that doing a CPU comparison involving H.264 would be something interesting to do. My only question, other than Quicktime 7 and Nero, what are you folks using to encode H.264 on the PC?
Remember Gigabyte's i-RAM from Computex? Well, one should be in my hands very soon and given the interest in it, it's going to receive top priority as soon as I've got it. Which begs the question, are there any particular tests you all would like to see? I'll admit, I am a bit surprised by the positive response the i-RAM received; I expected people to be interested in it, just not this interested in it.
Facebook
Twitter
RSS
55 Comments
View All Comments
xtknight - Friday, July 15, 2005 - link
I use x264 for H264 encoding. I find it gives the best results...much better than MainConcept's H264 encoder.Mike - Friday, July 15, 2005 - link
New system that I just received the parts and put together has a problem.Any suggestions on a DFI LANPARTY UT nF4 SLI-DR and Corsair TWINX1024-3200XLPT 1GB Kit DDR400 XMS3200 Dual-Channel Xtra-Low Latency Memory w/Platinum Heat Spreader problem? The problem is when both DIMMs are in the motherboard it reboots about the time Windows XP is ready to display the account screen. Each DIMM works fine when in one at a time. The mem86 test fails on Test #8 when both DIMMs are in, but passes when one at a time is put on the motherboard. I have tried the recommended 2.75V setting and less, but same problem occurs. Do you think I have a setting problem or a motherboard problem or a memory problem? Please let me know if you think it is a settings problem that I can try as soon as possible. Thank you.
TheChefO - Friday, July 15, 2005 - link
Joe-I see your point. However, I'm sugesting a regularly updated piece which tracks the progress of developers and their findings. Including their work on prerelrease hardware. Many devs have code up and running on this hardware and are merely tweaking their code for release on the dev kits. MS released devkits this month which are based on the same hardware as the final x360 but which run at lower clock speed. AFAIK Sony has done the same recently. This info from Devs would not be a representation of what to expect come launch time for either console, it would be more of a journal of the progress and accomplishments of these developers.
Joe - Friday, July 15, 2005 - link
TheChef0-While it's true the GPUs could be more easily disected at the current point, the CPUs will take a good deal of time for the "hundreds/thousands of devs out there that are busting their butt" to approach the point of full control.
I'm not trying to say there should neve rbe such a discussion with various devs, but merely that at the current time, with beta kits out for one system and possibly alpha kits out for the other, few to no devs will have any real information on what truely can be done. Almost all of it will be hearsay and speculation.
Once we get to the point that games enter the gold state, or very close to that point, we can possibly start getting honest information from them. Until that point, however, most of what devs will give will be based on speculation.
I'd rather wait on such an article and get worthwhile information than have it come out now and be based on possibilities. Once Xbox360 hits the point where the hardware is 100% nailed down and developers have used final state dev kits, I'll start taking their word for more than I do now. Similarly, I will take any comparsion between Xbox360 and PS3 with a grain of salt until the PS3 has its hardware specs nailed down. Already some of the touted features of PS3 have been dropped, and it's easily possible more will follow, even some of the "important" ones.
I say wait it out. Wait until there's the 100% correct information out there, then compare. Anything beforehand, even from "inside sources," will still be biased.
TheChefO - Friday, July 15, 2005 - link
Joe-True most devs have said they are not attempting to release games within the next 12mos or so that are multithreaded, but this does not touch on the developers that are looking to implement multithreading for a release further down the road.
Also any dev that is looking to deliver a game within the next 12mos is still using the gpus in the ps3 and x360 and therefore is able to answer questions regarding poly throughput, fill rate, etc. I would love to have an insiders look into these devs and I think Anand has the connections, know-how and means to get that info to his readership.
There are a lot of unanswered questions and a lot of interest in this segment which could be a boon for anandtech and supply the readership with honest info striaght from the devs mouths.
This would obviously also be a resource for the devs who up to this point have been largely muted without a voice to the media in any real measurable way. The majority of the time if you are hearing about a game it is either through an hr person from some publisher with an agenda to sell something or its Tim Sweeny/John Carmack. What about the other hundreds/thousands of devs out there that are busting their but just as much and in their own way have created innovation in specific ways for their specific projects. How about someone talking about proceedural synthesis on x360. Or how some dev has figured out how to get the cell to operate on 20k bones in their latest breakthrough in their physics engine.
There is a desire to know the innards of this fascinating world which we glimps in small pieces every once in a while and I think Anand could make this a beautiful regularly updated piece which could be called the developers corner which could go over the next generation games not only on console development but also on the pc. With the advent of the ppu, dual core cpus, and knocking on the door of photrealism in gpus we are in a very exciting timeframe which Anand could walk us through.
Is this possible Anand?
Joe - Friday, July 15, 2005 - link
I'd have to say, I don't believe that getting information from many of the current developers would help tremendously.Look at the suppositions that most, if not all, near-future games for the systems will not come close to taking advantage of either system.
The only team I've heard of that even attempts to claim they are pushing the system (Xbox360 in particular) to its limits would be Team Ninja doing Dead or Alive. The lead there claimed that his team was spending tremendous amounts of time playing with the multithreading power of the system to make the game as powerful as possible on that end.
With the idea that so few will be able to push them to the limits at this point in time, I don't see why any developer's word should be taken too highly. It seems like most so far have been vouching for each system based on the theory, rather than the practice. This holds particularly true for the PS3, which as far as I've seen does not have development kits out that even come close to the final result.
TheChefO - Friday, July 15, 2005 - link
Anand!Here I've been stewing with conspiracy theories in my head the past week and you pop up in your blog and throw them all out. Well glad to hear things are ok and in fact we will be privy to more insider info on the next gen machines.
Personally, heres what I'd like to see...
A compilation of quotes from anonymous developers giving specific numbers for what they are able to achieve with the hardware they have in house and a loose description on what that hardware is (alpha, beta, etc). The specific numbers would be fill rate, realistic aa levels at playable rates, polygons/sec, texture sizes, new techniques on physics/ai.
Basicly a large sampling from many dev's working on the hardware and giving their specific input on these categories and preferably contrast with the last generation (xbox, ps2, gamecube).
This would put things in a much better perspective on what we can expect with this generation and would be even better if you could do this as a series which could be updated on a quarterly basis and give some feedback on progress that these devs are making with this new hardware all while keeping their specific quotes without identifying them specificly.
This would be a great insight into this magnificint black art of pushing unknown hardware into unproven and exciting directions.
If this would not be possible, could you be so kind as to explain why? And if not would it be possible to get some semblence of what I described?
Thanks Anand!
Anonymous - Friday, July 15, 2005 - link
[emote] haha [/emote]Anonymous - Friday, July 15, 2005 - link
[emote] haha [\emote]mariush - Friday, July 15, 2005 - link
off-topic : Welcome Back, Anand ... I missed you..