First Socket FM2 APUs Spotted in Russian Store

[Atom_Anti]

Take any CPU limited-game?

Iguess what I'm saying is that asking what benchmarks people want to see may not exactly fit the bill...you may under up running a bunch of useless benchmarks that show no differences at all...

My test based on what people are use commonly. I could include more games, but what difference will make more titles if 7 heavy title already can tell Trinity is much faster for gaming than Llano or Ivy Bridge and whatever is the CPU performance!

About CPU tests? That is it! What else CPU tasks people are using more commonly then compressing files and encoding video?

 

[cadaveca]

For testing the gaming performance of the APU, it's going to be difficult to seperate the affect the CPU is having on the GPU by benchmarking just ingame FPS. to isolate the GPU portion use 3D Mark 2011.

I do not agree. We can discusss the "ins and outs" of why in the near future. I need to generate soem benchmarks to show why the exact opposite gets shown by 3DMark 11.

My test based on what people are use commonly. I could include more games, but what difference will make more titles if 7 heavy title already can tell Trinity is much faster for gaming than Llano or Ivy Bridge and whatever is the CPU performance!

About CPU tests? That is it! What else CPU tasks people are using more commonly then compressing files and encoding video?

I hear what you are saying. But, there are several billion people on this planet, and so there are several billion different opinions of what is important. You jsut need to show enough to appeal to some of those opinions, just keep in mind that the less benchmarks you include, the less opinions are covered...and the more peopel aren't goning to be satisfied with what they see.

I'd love to post every test possible in my reviews, but time does not allow. So you gotta pick those that seem worthwhile to you, and go with it. I run about 20 tests, and post scores from about 12-15.

 

[Atom_Anti]

For CPU-dependant synthetics, I like to see a varied range. Between 10-20 different benchmarks. Wprime, Cinebench, Pass Mark Vantage: Productivity score, 7Zip, WinRar, TrueCrypt, x264 HD, Handbrake. Just to name a few. There are many more good ones.

Another important thing is to do each benchmark twice. Once in single threaded mode and once in multi threaded mode. (usually in the options you can customise the threads/cores utility)

For testing the gaming performance of the APU, it's going to be difficult to seperate the affect the CPU is having on the GPU by benchmarking just ingame FPS. to isolate the GPU portion use 3D Mark 2011 and Heaven DX11.

Basically you are talking about video encoding and compressing files with different programs. Fine, in the future I will consider to do the same-tasks with different programs.

Believe me, I've run tests even 3 times, just to make sure... However to measure the power of an 4 core APU, I've rather test all the 4 core performance than just one Core. Today's and future's programs default to use all available CPU cores, later GPU cores too (called OpenCL acceleration).

See 3DMark11 in my test .

 

[Dent1]

but what difference will make more titles if 7 heavy title already can tell Trinity is much faster for gaming

It shows that you are thorough and it adds credibility to the review.

I do not agree. We can discusss the "ins and outs" of why in the near future. I need to generate soem benchmarks to show why the exact opposite gets shown by 3DMark 11.

Fair enough, not everyone will agree with 3D Mark 11. But I feel it should be included for honorary reasons atleast. It's a traditional, PC enthusiasts need to see 3Dmark otherwise the review seems incomplete.

See 3DMark11 in my test .

But it wasn't clear whether the results were the GPU score, CPU score, Physic score or overall score. You need to compartmentalise the results.

Overall, I feel you reviewed the GPU better than the CPU portion of the APU.

I've rather test all the 4 core performance than just one Core. Today's and future's programs default to use all available CPU cores, later GPU cores too (called OpenCL acceleration).

Lets say I was unsure whether to buy a dual core Trinity A4-5300 or a quad core Trinity A8-5500 and I want to do encoding.

I need to justify the spend. So I need to know the separation gap between a multi threaded and single threaded application or what type of boost I will get.

 

[Atom_Anti]

Every other commercial reviewer from Tom's Hardware, Hexus, Trusted Reviewer, Hot Hardware, Legit review and many others all show the Trinity 4 core outperforming the Llano 4 core both in gaming and non gaming tasks.

Of course, as I mentioned above they were using stock clocked A8-3500M. If I underclock my A8-3550MX to 1.5GHz, than I will also end up with lower results than Trinity.

But it wasn't clear whether the results were the GPU score, CPU score, Physic score or overall score. You need to compartmentalise the results.

Overall, I feel you reviewed the GPU better than the CPU portion of the APU.

For 3DMark11 I always mark overall score as most of reviews.

My review is certainly different as what we used to seen before. Basically I gave as much for CPU part as what it matters in real world situations. CPU performance difference sensible during movie encoding, compressing folders, but loading internet browser, Word, Power Point even the very slow A6 Trinity seems as fast as an overclocked A8 Llano or Core i5. Perhaps CPU have effect for games, but GPU performance will matters lot more... Here is a fresh good read about CPU effect for games.
Let's say I just do not want confuse people with bunch of CPU tests, what they won't ever use. 90% of the consumer ain't run any heavy CPU-task, but they still think CPU performance is the real deal in computers because all the reviews are highlighting the CPU performance in many style.
By the way if I'm gonna run more CPU tests, the results will be very similar; 2.4GHz Llano will beat Mobile Trinity in every CPU task, single-thread situation too.

 

[tacosRcool]

AMD is being a sneaky snake