AMD Benchmarks Zacate APU, 2x Faster GPU Performance than Core i5

[Manicdan]

An interesting feature would be if the CPU throttled down to match the performance of the GPU. So if the GPU is capable of 60FPS and the CPU is capable of 130FPS, it could throttle down to save power without drop in framerate. Some of that power could be used to boost the GPU. A frame limiter at say 60FPS controlling the turbo and throttle would save even more.

hopefully it can be smart too. in WoW your very cpu limited, so if you know to OC one core, leave the second stock, turn off the rest, and maybe adjust gpu until both one thread and the gpu are at the same usage (like 100% load on both) that would really make things easy. best framerate per watt ratio ever.

[JumpingJack]

Apparently AMD are using a different version of ie 9 preview that is available from the public, (presumably the developer one). That pretty much renders all the browser tests invalid.



- Fry on S|A

Nope, they weren't using the correct driver...

http://www.anandtech.com/show/3933/a...ormance-update

[Hornet331]

So there's only two types of games in your mind. Those you can play on an IGP from 4 yrs ago, and those you can play on a minimum of Llano class GPU?

Nope, playing games at minimum or sub 30fps is in my books not playing games... its a cramp.

With Llano type cores your able to play most games at medium with da decent resolution. Thats playing a game.

[duploxxx]

If AT had tested batman with Low Quality that would have been interesting and directly comparable to SNB preview results.

EDIT: Tried the N-Body Simulation on my friends desktop with a GTX 480 i am getting around 600-660 GFLOPS after which it crashes so i guess 23 GFLOPS is not bad for such a tiny little thing.

don't forget mobile parts will have lower clock and culv will have even less base clock/turbo.... and there are GT1 and GT2 versions.

according to anandtech latest update this was the full blown GT2 with lower clocks and it was more or less on par (some win some loose) with 5450 so 80SP 650mhz

[Mechromancer]

Nope, they weren't using the correct driver...

http://www.anandtech.com/show/3933/a...ormance-update

Yes, AMD wasn't using Intel's newest driver but they allowed Anandtech to manually install the newest driver and another game to test with.

Anandtech:

It’s very rare for any public company to make an on the spot decision to let us benchmark and publish test data of an unreleased part without having ever seen it before. The first time the AMDers in the suite saw Zacate running Batman was when we installed it. To be honest, it was probably the most open and flexible I’ve ever seen AMD be. I knew if the IE9 numbers changed that it would call the City of Heroes numbers into question. By allowing us to rerun everything as well as add an additional title (one that we’ve used more recently) AMD handled the situation perfectly.

For a chip going into a sub-$500 notebooks/netbooks this little guy isn't too bad. It's not like people will be Folding@Home and doing BigADV WUs on one of these things. CPU power isn't quite as important as the ability to accelerate media and do some casual gaming.

[Particle]

I'm really excited overall to see integrated graphics going this way. Considering how nice the AMD IGPs have been compared to what came before and now to see how much what is coming next outclasses the lustrous products of today, I think computing in general is about to become a lot nicer. It even effects us, since we're the people Joe Random-with-a-computer-that-has-integrated-graphics calls upon to fix their computer when they run "Free Tacos, Registry Cleaner, & Happy Fun Cursors.exe" from totallylegitvirusfreedownloads.com which they in turn found by clicking on a banner ad at turboclown:banana::banana::banana::banana:.org. They'll at least have a responsive UI.

[richierich]

That's great how AT was able to resolve the driver issue and compare another game. Can't wait for the new platform to release.

[blindbox]

There goes all doubts. This is great! Still, I'd rather see cpu performance (we know how the gpu's going to perform anyway). If all goes well... well... I want to host a server on one.

[Xoulz]

fusion is a winner....

[Lmui]

That is an insanely powerful chip for its specs IMO. It looks like fusion is going to be in my next laptop for university a few years down the road unless intel beats them with SB or whatever's after SB. To be completely honest, in my experience, at university, there are three uses for laptops, notetaking/reading/research (easily adequate), light gaming (adequate), and for some specific people, heavy CPU usage programs such as rendering which is the only thing I'm not too sure on for fusion. If the program supports using the GPU to assist the CPU for stuff like that then AMD just got me to buy something other than their graphics cards.

[Nintendork]

There goes all doubts. This is great! Still, I'd rather see cpu performance (we know how the gpu's going to perform anyway). If all goes well... well... I want to host a server on one.

Once you get rid of the anemic perfomance from Atoms (wich Athlon II / Turion II neo shows the path) the next thing is core frequency and IGP.

Current Turion II X2 Neo K625/K665 1.5/1.7Ghz @15w should be a great choice for a home server with 64bit and virtualization support). What Ontario/Zacate will bring is the lower consumption and high frequencies for the top of the line.

Say Zacate with 1.8Ghz-2Ghz dual core.

[JumpingJack]

Yes, AMD wasn't using Intel's newest driver but they allowed Anandtech to manually install the newest driver and another game to test with.

Anandtech:


For a chip going into a sub-$500 notebooks/netbooks this little guy isn't too bad. It's not like people will be Folding@Home and doing BigADV WUs on one of these things. CPU power isn't quite as important as the ability to accelerate media and do some casual gaming.

Yep. AMD has done a first class job with this one netbooks are going to be very different in a few months. It's better than just casual gaming I can casual game on an Atom netbook. (plants v zombies, Luxor, etc) but here we can get to some gen -1 type action games.

[Hornet331]

don't forget mobile parts will have lower clock and culv will have even less base clock/turbo.... and there are GT1 and GT2 versions.

according to anandtech latest update this was the full blown GT2 with lower clocks and it was more or less on par (some win some loose) with 5450 so 80SP 650mhz

Mobile SB are GT2 only and have same turbo clocks as the desktop moddels. We know nothing about LV. If they follow the same trend as with the pervioues i5s we get the baseclock of the normal mobiles as turbo mode for the LV models.

Which likely will be 650mhz. It will be slower what anand has shown... or not, depending on how much they can extract from drivers.

[god_43]

well at least amd redeemed themselves, by letting anand play with the test.

[informal]

^^And Ontario still reigns superior in games,even in this early stage of clocks,BIOS,drivers etc. When it launches on the market,I bet we will see even better performance and lower power.

[kl0012]

Have tried Bathman demo on my I3-530 with default GPU clock (733MHz)
The FPS varies between 12 to 22 with average ~14-15 FPS.
CoH is still downloading.

[god_43]

Have tried Bathman demo on my I3-530 with default GPU clock (733MHz)
The FPS varies between 12 to 22 with average ~14-15 FPS.
CoH is still downloading.

lol still? well hey, you raised the flag when no one else saw it, gj. turns out it was a mistake but anand cleared it up.

[kl0012]

Here is the same position like in Anand's test. So his test seems valid, however i dont think i5 gpu was running 766MHz.

[Nintendork]

The Zacate of the preview is running at 1.6Ghz on the bobcat core (by fudo). And obviously clock per clock are slower than i5/i3.

A nice idea can be comparing this Zacate to a Turion II neo 1.5Ghz with HD4225. Both bottlenecked by cpu frequency. And Zacate with just single channel.

[ajaidev]

Here is the same position like in Anand's test. So his test seems valid, however i dont think i5 gpu was running 766MHz.

The difference is 1FPS between yours and AT's test which can be left to error margin or other factors.

[god_43]

Here is the same position like in Anand's test. So his test seems valid, however i dont think i5 gpu was running 766MHz.

maybe they didnt set all the gb's?


;p.

[kl0012]

The difference is 1FPS between yours and AT's test which can be left to error margin or other factors.

Possible. Also it might be because of slower mem speed in notebook/slower cpu freq.

[railer]

happy to see it was a fluke or else i would loose respect for AMD. Impressive performance for the size and wattage. In 8 years we wont need gpu's for gaming it be all APU's. My last name is nosterdamus btwy.

[flippin_waffles]

Bullet Physics demo here:

http://pcper.com/article.php?aid=1003

The demo itself wasn't particularly exciting but the devil is in the details. Zecate was running a DX11-only, Shader Model 5.0 cloth simulation that will be part of the upcoming Bullet Physics, the open-source physics engine for gaming and 3D rendering. The Intel Core i5-520M based notebook on display obviously couldn't even open the demo and all early indications are that Sandy Bridge will not be a DX11-ready GPU either, giving AMD's APUs an advantage in a specific area through 2011.

[tajoh111]

You guys are giving AMD way too much slack. It should have been obvious that something was a miss with the initial benchmark used and AMD was obviously trying to pawn off their performance or inflate it higher than it actually is. I don't think AMD is incompetent enough that they couldn't get simple driver things resolved on a test platform for a comparison.

They just happen to let a driver issue get by that happened to inflate their performance?

People are being paid money when they set up these things, it an honest mistake when it joe shmoe, but when a professional company does it. They should be making mistakes like that.

Some of you guys piss on review websites that don't uses updated drivers in reviews. This is even worse than that because this info(driver dates) are not readily available where in most reviews, they tell you exactly what drivers they used.

Basically anandtech caught them cheating and they had to back up on their words and give anand freedom to do what he wanted.

If they didn't, Anand would have published that AMD was cheating on his website and that would have been bad press, even with AMD users.