The GPU transcoding won't load CPU cores much at all. They should be clos to Idle. And you can see the opposite on the Intel. CPU consumes more than GPU on the Intel in it's specific test, while GPU consumes more on Llano.Originally Posted by TESKATLIPOKA
Sigh.
AMD A8-3850
Intel Core i3-2105
Idle
43.6W
51.7W
GPU Accelerated Video Transcoding
126W
85W
3D Gaming (Metro 2033)
126W
101W
CPU Load (x264 Encode)
123W
87.6W
Taxing the cpu = 123-43W = +80W!! (gpu is idle in both cases!)
Gaming = 126W-43W = +83W where gpu and cpu are used intensively. (There is no turboboost!, cpu runs at 2.9GHz).
Transcoding = "". high gpu load with load on cpu's (cpu is not idle and is running at 2.9GHz).
And yes you can compare those TDP values. Go read techreport whom tested a 3800 and had similar or lower powerconsumptions than the i3 for cpu load. (and not 36W higher)
How do we have to read this? Is this the scenario like with the rad69xx and nvidia590 cards where the drivers delivered were still buggy? Do you imply there are still driver issues, or simply that their is lack of support in alot of test programs. (aida, sisoft, ..)
ok so bottom line the launch platform is still having issues. So is it fair to say they have put more effort in creating a nice box than they did in making sure the content is what you expect?
seems nothing has changed over the years, what a shame ...
anyway, thanks for posting your impressions here.
Power Rig: Core i7-5930K, ASRock X99 Extreme6/3.1, 16GB G.Skill DDR4-2400, Asus Strix GTX980 OC
Time Sink: Core i7-5775C, ASRock Z97E-ITX/ac, 16GB AMD DDR3-2133, Silverstone PT-09 w/ 120W Power Brick
HTPC: Athlon 5350, ASRock AM1H-ITX, 4GB DDR3, Supermicro SC-101i
What Bios version and Board are you using?
-Boris-
1) I was talking about MetroThe GPU transcoding won't load CPU cores much at all. They should be clos to Idle. And you can see the opposite on the Intel. CPU consumes more than GPU on the Intel in it's specific test, while GPU consumes more on Llano.
2)
idle 43.6W
GPU trans-coding 126W the difference is 82W and the same discrete gpu with a bit higher clock as the IGP has a max. TDP of 39W 82-39=43W just for CPU if IGP is working at its TDP limit. So much for your statement how cpu is close to idle.
3) SB consumes more in Metro than in x264 Encodeand if you don't know SB has Sandy Bridge Media Engine which is doing video encoding and decoding instead of GPU thats why the consumption is lower than in x264 Encode.
I'm not even sure what the argument is anymore, all I can say with certainty is:
- It's the CPU cores that draw the most power by far under 100% load. As mentioned almost all the TDP is eaten up by the CPU.
- If you artificially load up both the CPU and GPU you will far exceed the TDP and the system will overheat and shutdown on a std heatsink (I tried it).
This doesn't happen in real life gaming (even using all cores) because both the CPU and GPU fight for memory BW, rather than maxing out their execution resources.
- The GPU is on the CPU-NB power plane, and it's the CPU's stock Vcore that's pumped through the roof
I have figures from my own testing on my laptop i'll dig them out
Assume? Why don't you monitor it?
Mine? I don't have this press kit, or a single ES.Could you link me to your driver version ?
I was referring to what JF said:
From here: http://www.overclock.net/amd-cpus/11...aunch-faq.htmlQ. I saw a benchmark on xyz website. Is that how bulldozer will perform?
A. No. Nothing posted before launch will be representative of actual performance. To get actual performance, you need:
Final production silicon
Final processor microcode
Final system BIOS
Final OS optimizaitons
Final drivers
An app compiled with the latest flags
A person who understands the app and configures the test properly
Although, I'm not sure if he was referring to a certain CPU driver, as well, or only those for rest of ther system.
I know C'n'Q needed a driver (or a 3rd party app). I don't know what's the case with f.ex. PowerNow! 3.0 and TurboCORE. Perhaps it's all controlled in HW already. Better ask JF about it...
(Searching for CPU drivers on AMD's site I've found only old ones.)
One very interesting post from SA. We now have a new slide deck which is different from the one leaked earlier. Especially the slide covering FX vs i5 and i7.
Has anyone noticed that between this slide:
And this slide:
The 2600K and FX have been swapped around showing the same difference in numbers? The second slide makes more sense with the results being order of the key. Whoever made the first slides decided it would be a good idea to swap 2600K and FX performance...
It would make sense, because the order in the legend is now the same as the order of the bars. I never noticed that before.
Let's see what the latest slide shows us:
8 tests in total.
We can say that Zambezi FX wins hands down in at least 4 or even 5 of them (Handbrake is a win for FX but by a few percents;still a win).
It looses in one test(7 zip) by a small margin (few percents again).
It loses by a decent/big margin in 1 benchmark (wprime) and it loses by a smaller margin in also one benchmark which is Bibble (by 5 to 7% tops).
New slide deck shows us that in following benchmarks FX is NOW faster than 2600K,instead of being slower as depicted in earlier version: Winrar4.x264 pass2,pov ray,abbyy OCR. In wprime FX is NOW slower than 2600K and this is the only case where new slide deck shows FX being slower versus the old one.
All summed up:According to AMD's newest slide deck, FX XXXX @ XXGhz is overall somewhat faster than 2600K. This is not what all the leaked benchmarks are showing us though.
Last edited by informal; 10-07-2011 at 03:52 AM.
What I hope that after bulldozer launches, somebody would do the following:
Disable some modules of the 8150 and search the optimal performance
(eg. 1 module disabled, 2 cores disabled.. and then MEGA oc no 6xxx and 4xxx series because you want to do this with the best cores/modules)
Lesser modules means more cache per core more mem bandwith per core and less overheard, no?
And offcourse, SB with HT off vs bd with 4 threads
Last edited by donitsi; 10-07-2011 at 04:11 AM.
Why focus on gaming, both GPU and CPU are loaded in gaming? It doesn't show anything. Compare GPU only(mostly) test with the CPU only(mostly) test. All you can tell is that in this particular case the GPU test loads the processor more than the CPU test. You really can't turn those figures to that it's the CPU that is the power hog. If you have other reasons for those claims post them instead. The figures you have there doesn't cut it to prove your point. And you can never compare power consumption between different reviews at different sites with different boards. So that's just another bad conclusion.
I do believe you guys might have a point, but it seems to that you haven't presented it here, the comparison above just doesn't prove your claims.
1. Yes, and metro doesn't show-Boris-
1) I was talking about Metro
2)
idle 43.6W
GPU trans-coding 126W the difference is 82W and the same discrete gpu with a bit higher clock as the IGP has a max. TDP of 39W 82-39=43W just for CPU if IGP is working at its TDP limit. So much for your statement how cpu is close to idle
3) SB consumes more in Metro than in x264 Encodeas it loads both CPU and GPU. That's why I said you should compare GPU test vs. CPU test.
2. Where do you have these numbers of internal TDP limits for different parts?
3. Still, in this particular test GPU test loads the processor harder than the CPU test.
No because people forget some arguments and focus on other stuff, brings numbers that actually don't say anything thus derailing the discussion from the subject. The argument is, you can't predict Phenom behaviour on 32nm with Llano as a base. There is way to important differences for that to be possible.
You actually bring some points here. And I don't doubt that I might be wrong. You can please dig them out. The other arguments I've got this far haven't showed anything conclusive.
But my main point (which everyone forgets) is that Llano probably isn't made the same way as a Phenom II would be made at all. AMD talks about differences in silicon between different product lines, like Turion and Athlon/Phenom, and between Thuban and Deneb. It's clear that some design choices that might be good for a GPU isn't as good for a CPU. Low power high density trannies might be good for lots of units packed in a GPU at sub 1GHz frequencies. But might need more current to work good in a 3GHz CPU.
One thing more, I don't say that Thuban will be close to BD, it's just an if. If Thuban is close to bd, then I think a 32nm Thuban would be better of.
Last edited by -Boris-; 10-07-2011 at 04:17 AM.
if I compare 4.2 (turbo) vs 4.0 ghz (locked) the single threaded scores are scaling correctly. I never leave extra apps open when testing... Turbo works don't worry... but it won't make up for what I see here in single threaded apps...
Only driver I now off were the new CAT preview drivers... no other CPU, CNQ or such in my posession... or that I know off that need to be installed.
Question : Why do some overclockers switch into d*ckmode when money is involved
Remark : They call me Pro AsusSaaya yupp, I agree
The same people that, watching the first slide, claimed that i7 2600k was better and that wprime 32m did not matter, now, that the results are inverted, what do they say now??
Watchig the second slide (the correct one I think, if it cames from amd.. the other slides were faked by someone, if you look at the first slide and the first benchmark you can see clearly that it is edited) they will say that i7 2600k is still better and that wprime 32m is the most important test.
LOL
asus Crosshair V Formula tested 0083 0813 and 9905 bios now (last one gives best performance and opens ram dividers to 2400)
Question : Why do some overclockers switch into d*ckmode when money is involved
Remark : They call me Pro Asus
yes, this is probably the one thing that prevents a proper comparison. It may be the result of porting Evergreen architecture over, I don't know. However, whilst Intels GPU isn't really comparable, it's still a GPU and they don't seem to have any such issue at all, so IF what you say was the case, I'd hope for AMD's sake there's no such issue for trinity.
-Boris-Actually I did that too because Metro and GPU acceleration have the same power consumption1. Yes, and metro doesn't show ....... as it loads both CPU and GPU. That's why I said you should compare GPU test vs. CPU test.
2. Where do you have these numbers of internal TDP limits for different parts?
3. Still, in this particular test GPU test loads the processor harder than the CPU test.
Thinking cpu is idling during gpu acceleration is very stupidor are you saying that media chip is consuming 44W
.
Your argumentation was totally flawed from the beginning, The End
P.S. Did I tell you that TDP 39W was for a similar gpu built on 40nm and not 32nm and memory chips should have been included in said TDP too so IGP in load will have <<39W
you are seriously hilarious, can't even admit you are simply wrong because my TDP comment is more than enough to destroy anything you bring up to defend your claim how IGP in Llano is consuming more than cpu.Why focus on gaming, both GPU and CPU are loaded in gaming? It doesn't show anything. Compare GPU only(mostly) test with the CPU only(mostly) test. All you can tell is that in this particular case the GPU test loads the processor more than the CPU test. You really can't turn those figures to that it's the CPU that is the power hog. If you have other reasons for those claims post them instead. The figures you have there doesn't cut it to prove your point. And you can never compare power consumption between different reviews at different sites with different boards. So that's just another bad conclusion.
I do believe you guys might have a point, but it seems to that you haven't presented it here, the comparison above just doesn't prove your claims.
Last edited by TESKATLIPOKA; 10-07-2011 at 06:40 AM.
here another slides:
FX-8150_slide1.jpgFX-8150_slide2.jpgFX-8150_slide4.jpg
do you think the FX 8150 is competitive when compare to i7 2600k?
fwiw, cache sizes on the 8150/8120/8100 are 8MB L2 and 8MB L3 (16MB total).
cache sizes on the 6100 are 6MB L2 and 8MB L3 (14MB total)
cache sizes on the 4170/4100 are 4MB L2 and 8MB L3 (12MB total)
that should cut some rumour mongering out.
dave
Heat: 50 - 0 - 0 under "Argus333"
note to reviewers:
Anyone who uses bd with crappier memory than 1866, will get smacked to the face
Even the puny llano 3850 gets a small perf boost on 1866 memory (cpu part)
And what exactly are the issues you have found in bios 9905?
The review kit with proper CPU-Z, HWMon, HWInfo, AIDA is going to be released later today.
mhm... would i like to review right now or should i be happy other people do it
question
1. ASUS Sabertooth 990fx | FX 8320 || 2. DFI DK 790FXB-M3H5 | X4 810
8GB Samsung 30nm DDR3-2000 9-10-10-28 || 4GB PSC DDR3-1333 6-7-6-21
Corsair TX750W | Sapphire 6970 2GB || BeQuiet PurePower 450w | HD 4850
EK Supreme | AC aquagratix | Laing Pro | MoRa 2 || Aircooled
Posting Permissions
Reply With Quote
MAXIMUS IV EXTREME-Z CPU INTEL 2600K @ 4700 Mhz @ 1,296v liquid cooling by EK supreme hf RAM A-DATA XPG+ V.2 2200 C8 elpida hyper SLI MSI GTX 560ti TWIN FROZRII OC liquid cooling by ek supreme hf vga bridge edition SSD CRUCIAL C300 128GB MONITOR SAMSUNG 2233RZ + NVIDIA 3D VISION KEYBOARD LOGITECH G110 MOUSE LOGITECH MX518
Bookmarks