Forgot your password?
typodupeerror
AMD Graphics Intel Upgrades Games Hardware

AMD Trinity APUs Stack Up Well To Intel's Core 3 223

Posted by timothy
from the those-ndas-must-have-weighed-a-ton! dept.
Barence writes "AMD's APUs combine processor and graphics core in the same chip. Its latest Trinity chips are more powerful than ever, thanks to current-generation Radeon graphics and the same processing cores as AMD's full-fat FX processors. They're designed to take down Intel's Core i3 chips, and the first application and gaming benchmarks are out. With a slight improvement in applications and much more so in games, they're a genuine alternative to the Core i3." MojoKid writes with Hot Hardware's review, which also says the new AMD systems "[look] solid in gaming and multimedia benchmarks, writing "the CPU cores clock in at 3.8GHz / 4.2GHz for the A10-5800K and 3.6GHz / 3.9GHz for A8-5600K, taking into account base and maximum turbo speeds, while the graphics cores scale up to 800MHz for the top A10 chip."
This discussion has been archived. No new comments can be posted.

AMD Trinity APUs Stack Up Well To Intel's Core 3

Comments Filter:
  • by IYagami (136831) on Thursday September 27, 2012 @10:39AM (#41477585)

    AMD allowed websites to publish a preview of the benchmarks before the estimated date if they only focused on graphics performance. This is an unfair move by AMD.

    Read http://techreport.com/blog/23638/amd-attempts-to-shape-review-content-with-staged-release-of-info [techreport.com] for more details

    (maybe in a couple of weeks you will find that AMD Trinity APUs have abysmal x86 performance compared to Intel CPUs)

    Disclaimer: I own a laptop with an AMD cpu inside

  • Re:Wow (Score:2, Informative)

    by TeXMaster (593524) on Thursday September 27, 2012 @12:16PM (#41478805)

    Or use Windows or possibly Gnome...or do OpenCl or OpenGl programming...or-

    The list goes on. The fact that people are still selling craptacular integrated video chipsets in this day and age saddens me greatly. Guys, it's 2012...pony up for a dedicated video card with dedicated video ram. Quit trying to save a buck or two on a component you really don't want to be cheap on.

    Well, I think you can do OpenCL on Intel HD3xxx/4xxx chips these days.

    AFAIK, Intel HD3xxx is not OpenCL capable, and Intel HD4xxx is officially supported by Intel on Windows only (no Linux drivers). This is in sharp contrast with AMD, which has much better OpenCL support for everything they ship (CPUs, GPUs and APUs).

  • Re:Wow (Score:1, Informative)

    by hairyfeet (841228) <bassbeast1968@gma i l . com> on Thursday September 27, 2012 @12:53PM (#41479267) Journal

    Because its adding heat for a part you're not using and sucking up die space?

    That said if Trinity is based on Faildozer, which I believe it is, you have to ask yourself ONE question....do you like WIndows 8? If the answer to that question is no? You might want to avoid it as ONLY Windows 8 has the scheduler bugfix. For those that don't know the Bulldozer/Piledriver design uses "half cores" because.....well its a server chip, it really is. Its designed for integer heavy loads like you see in servers and to pack more cores into a die, because AMD has had problems getting the IPC per core up, they went with a "half core" where two cores which equals a "module" have to share a FP unit.

    The problem is the fact that AMD "lies" for want of a better word, to Windows and tells it that its real cores instead of hyperthreading. Lets say you have an 8 core/4 module unit with two related loads, a1 and a2, along with two unrelated loads, b1 and c1. The way it SHOULD be scheduled is a1 and a2 on module 1,b1 on module 3/4 and c1 on module5/6. Instead Windows will give you something like a1 and b1 on cores 1 and 2, a2 and c1 on cores 3 and 4. Now this is NOT the fault of MSFT and Windows, because Windows has NO way based on what the chip tells it to know that all the cores aren't equal so it treats them as full cores instead of modules that share resources.

    Now this seriously bums me out, as I believe in competition and have been building AMD exclusively for the past few years, ever since the OEM bribery and compiler scandals came out. But since MSFT has already said they aren't gonna backport the scheduler fix (they have only released a patch, since withdrawn I believe) and thanks to this boneheaded design you'll have a boat anchor tied on your system you really only have 4 choices, 1.-Disable half the cores, 1 per module, so you are basically only getting half of what you paid for but each core then has a full FP unit,2.- OC the living hell out of it to use speed to make up for the penalty, 3.-Stick with the AM3 Phenom II units, this has been what I've been doing as the Phenom II quads and hexacores are dirt cheap now and still have decent speed, or 4.-Don't buy AMD.

    AMD recently hired one of the head chip designers away from Apple, so lets hope he can put together a rock solid replacement to BD/PD or at least get AMD to see that selling a server chip in the consumer market is NOT the way to go. As you can see from the speeds they are hitting they are following #2 on my list and OCing the hell out of them to make up for the penalty which if you don't mind the extra heat and the louder cooling you'll need to deal with it? Then please buy AMD, we don't want to see Intel become a monopoly again, those were REALLY not good times. This is why I'm still selling AM3 units, for the average person a Phenom II quad or hexacore really will do anything they want to do with cycles left over.

    But I think its only right to let those that don't know you WILL take a speed hit when using Win 7 on a BD chip thanks to the scheduler bug. Its a real shame AMD didn't simply have the chip tell Windows it was hyperthreaded, while it wouldn't have been a perfect solution it would have been better than lying and saying all core are equal when they are not. In any case I hope they sell plenty of these and I'm personally looking forward to Bobcat II as I loved their Bobcat netbooks. Having a quad core in a 12 inch netbook that gets 6 hours+ on battery? Sweet.

  • by hairyfeet (841228) <bassbeast1968@gma i l . com> on Thursday September 27, 2012 @05:47PM (#41482913) Journal

    Oh really? I'll quote and highlight the relevant portion "On 17 April 2007, NVIDIA released the GeForce 8500 GT, 8600 GT, and 8600 GTS for the low-end to mid-range market".

    Now please note that the midrange was NOT the 8600 vanilla, or GT, but the GTS. Its a trick nvidia has done for years, they make several cards with the same name but with different letters, so people say "I have an 8600!" while actually having the low rent card. they've done this going back to Geforce 4, where they had the 4, the 4ti, and the MX4000 which while having a larger number was actually equal to a G3 series.

    So I'm sorry but you're wrong. I can see how you were confused, just look at how many flavors they made of the 7600 or 8600, I think you are looking at over a half a dozen variants of just those 2 numbers, VERY confusing to the consumer.

    This is why I prefer the number schema started when AMD took over ATI, as it couldn't be simpler. the first number is the series, 3xxx, 4xxx,5xxx, the second number is whether its low mid or high, 42xx-43xx for low,45xx-47xx for mid, 48xx-49xx for high, and the next to last number further splits the group, such as 4830 for the low end of the high cards in the 4xxx series, 4850 for the middle, 4870 for the high, and if its a 9 for the third number, like 4890, then its a dual chip design.

    Hell of a lot easier than trying to keep up with which is which on the Nvidia side, GS vs GT vs GTS vs GX vs Ti and so on.

It is contrary to reasoning to say that there is a vacuum or space in which there is absolutely nothing. -- Descartes

Working...