Another review for the list on the 1st page:
http://www.phoronix.com/scan.php?pag...ynnfield&num=1
This one is a review under Linux and results are quite surprising(no amd quad core used,but instead a X3 Deneb)
Another review for the list on the 1st page:
http://www.phoronix.com/scan.php?pag...ynnfield&num=1
This one is a review under Linux and results are quite surprising(no amd quad core used,but instead a X3 Deneb)
Work Rig: Asus x58 P6T Deluxe, i7 950 24x166 1.275v, BIX2/GTZ/D5
3x2048 GSkill pi Black DDR3 1600, Quadro 600
PCPower & Cooling Silencer 750, CM Stacker 810
Game Rig: Asus x58 P6T, i7 970 24x160 1.2v HT on, TRUE120
3x4096 GSkill DDR3 1600, PNY 660ti
PCPower & Cooling Silencer 750, CM Stacker 830
AMD Rig: Biostar TA790GX A2+, x4 940 16x200, stock hsf
2x2gb Patriot DDR2 800, PowerColor 4850
Corsair VX450
I posted another one with a 26 applications showing almost the same numbers.
BTW ,under linux,even x3 710(2.6Ghz) is holding its own while going against the new i5 and i7-look at phoronix review under ubuntu(that's real world workloads). That's 3 cores @ 2.6GHz against 4c/8t @ 2.66-3.46Ghz
Last edited by informal; 09-08-2009 at 09:02 AM.
Ok as it stands i think the performance in high bandwidth CF "Like 4870 x2 CF" will be:-
i7+x58>PhII+790>i5+P55
Coming Soon
For someone looking to build a screaming mid-range machine, it's hard to argue with i5 and P55. Tons of performance, value etc.
Also, the motherboard selection is fantastic!
Can't agree. The second diagram (which excludes games tested on high res with 9800GTX) actually shows that i5-750 overal faster then Pii-965.
Hm... I wonder why they used Ubuntu 9.10 Alpha release. Alpha releases usually includes many debug code and has unpredictable performance on different systems.BTW ,under linux,even x3 710(2.6Ghz) is holding its own while going against the new i5 and i7-look at phoronix review under ubuntu(that's real world workloads). That's 3 cores @ 2.6GHz against 4c/8t @ 2.66-3.46Ghz
Well we can go all day long finding errors in testing or test choices. 750 due to its high turbo mode looks faster than 965BE,but without it it's a bit slower.Also with turbo on,750 is a bit faster or on par with 860 i7 too... The difference between i5/965 and top of the line 870 with turbo and smt on is 16 percent in office and multimedia apps... No gaming involved. 16% is decent but nothing to write home about,especially if you consider the difference in price.
Phoronix website is quite professional all around. There is a reason they used that release.They tested the 920 i7 quite a long while ago and were quite pleased with its performance. You can always ask them though .
Last edited by informal; 09-08-2009 at 10:16 AM.
pfff Kyle tested Lynnfield under gaming @ 640x480 what a champ!
http://hardocp.com/article/2009/09/0...7_processors/5
Adobe is working on Flash Player support for 64-bit platforms as part of our ongoing commitment to the cross-platform compatibility of Flash Player. We expect to provide native support for 64-bit platforms in an upcoming release of Flash Player following the release of Flash Player 10.1.
So you suggest that car reviewers pull out engine out of two ton car and put it on bench table to give you acceleration performance torque ratio, fuel efficiency etc...
that page is about "Lynnfield under gaming", and people in 2009 don't game on SMARTPHONE resolution!!
Kyle, and obviosly you don't have clue about CPU role in gaming... if you did have clue you would know that true stress for CPU under gaming is with more detail!
The point is to find sweet spot that would stress CPU, and still not represent GPU bottleneck!
Ideal CPU under gaming conditions would be testing with Dual GPU (1 gig cards) setup (driver overhead comes as a burden for CPU) @ 1680x1050 Medium/High settings!
Adobe is working on Flash Player support for 64-bit platforms as part of our ongoing commitment to the cross-platform compatibility of Flash Player. We expect to provide native support for 64-bit platforms in an upcoming release of Flash Player following the release of Flash Player 10.1.
Last edited by ajaidev; 09-08-2009 at 11:17 AM.
Coming Soon
Well in real function that would make the cpu faster, once again turboboost is an integral function of the cpu now as much as speedstep to lower the speed. If people are looking at the review to decide which out of the box cpu is faster then it is what it is, what they see is what they get.
Do you expect them to clock the i5 at the same speed as a comparably priced AMD cpu to be fair...
Work Rig: Asus x58 P6T Deluxe, i7 950 24x166 1.275v, BIX2/GTZ/D5
3x2048 GSkill pi Black DDR3 1600, Quadro 600
PCPower & Cooling Silencer 750, CM Stacker 810
Game Rig: Asus x58 P6T, i7 970 24x160 1.2v HT on, TRUE120
3x4096 GSkill DDR3 1600, PNY 660ti
PCPower & Cooling Silencer 750, CM Stacker 830
AMD Rig: Biostar TA790GX A2+, x4 940 16x200, stock hsf
2x2gb Patriot DDR2 800, PowerColor 4850
Corsair VX450
I don't think it is even worth to discus that. I prefer to wait for official release of Ubuntu 9.10 (somewhere in October) with all bugs filtered out (especialy those affected new hardware). I see no point in testing cpus on different alpha version of unstable software.
Is this what you're looking for?
http://xbitlabs.com/articles/cpu/dis...0_9.html#sect0
Are you serious? As mentioned, he's not stressing the GPU. The lower the resolution, the more CPU bound the game is.
If it were up to you, you would probably want to test it with AA also.
Gaming with more detail is more stressful for the GPU, not so much the CPU. In case you didn't understand, he's trying to isolate the CPU performance.
Jeez... did you even read the text of the article?
I quote Kyle (bold added by me):
You may not find the information useful, others might. Personally, I don't care as I don't game, but putting down a testing procedure without at least some basic research is wrong in my opinion. Not only does Kyle state his goal very clearly and percisely, he also raises the necessary caveats on how the data can be used.. Again, if you don't find the data useful move on. However, I don't know of any other way to get the answer to the question he is raising. Which is, with as little GPU bottleneck as possible, just how fast can the CPU go on a particular game?As always, these benchmarks in no way represent real-world gameplay. They are all run at very low resolutions to try our best to remove the video card as a bottleneck. I will not hesitate to say that anyone spouting these types of framerate measurements as a true measuring tool in today’s climate is not servicing your needs or telling you the real truth.
The gaming tests below have been put together to focus on the processor power exhibited by each system. All the tests below consist of custom time demos built with stressing the CPU in mind.
hmm so i5 has made AMD's mid range processors incompetent once again. I wonder what AMD will do to counter this? I hope whatever they do is better for themselves and consumers as well.
[SIGPIC][/SIGPIC]
Once again: as the scene that's being rendered simpler - there's less work to be done for CPU - less geometry to feed GPU, less traficking over the PEG, less driver overhead to deal with...
I agree that there's a point in testing CPU's where GPU is becoming bottleneck and it shouldn't be passed for the sake of shadowing CPU performance, but going so low as Kyle did is ewen worse 'cos it misleads readers about true potential and usability of CPUs in gaming scenarios
ABSOLUTELY! If I'm spending serious money on new powerful HW I want to see how it performs on my big 30" TFT, and apparently here's the answer:If it were up to you, you would probably want to test it with AA also.
http://www.anandtech.com/cpuchipsets...spx?i=3634&p=9
2560x1600:
He don't have clue... or maybe does... anyhow he's misleading readers! If he wanted to stress out CPU in gaming he should have run software rendering of Unreal 1 and Quake 1 in 640x480!!Gaming with more detail is more stressful for the GPU, not so much the CPU. In case you didn't understand, he's trying to isolate the CPU performance.
Insulating CPU in trying to portrait gaming performance is by definition nonsense, but you can make of it something if you know how to do it, and I've suggested how! And as someone pointed xbit is using that approach, as well as anandetch... http://www.anandtech.com/cpuchipsets...spx?i=3634&p=8
but yeah Kyle is smartest of them all
Adobe is working on Flash Player support for 64-bit platforms as part of our ongoing commitment to the cross-platform compatibility of Flash Player. We expect to provide native support for 64-bit platforms in an upcoming release of Flash Player following the release of Flash Player 10.1.
How is anand misleading? At that res you get very GPU limited very quickly. Ok it's Crysis but you know the further down the ladder of graphical fidelity you go the more cpu intensive it is, hello Source engine, hello COD4/5! It's a strange world in that if you have a 30" you are better off spending your cash on graphic card upgrades rather than cpu upgrades. The smaller your monitor the more cpu makes a difference, the bigger it is the less it does. So at the highest res available to consumers, you're fine with a high clocking e8x00. The sweet spot is 1920x1200, the cross section where you get most bang for buck from both. I've been considering downgrading for some time, but hopefully the 5xxx series and GT300 will change the noodle.
who said anand is misleading? read again my post... anand's and xbit's approach is much more informative than {H}ocp's
the smaller the monitor - smaller is resolution so my advice to people with smaller monitors is go with AA as high as possible!
I'd agree with you that this and next year will establish 1080p as a mainstream resolution that's covered with a range from 22 'till 26 inch monitors, and that's a resolution that should be targeted in reviewing!
When it comes to methodology, my POV is to tune down any Shader, or other special effects, to leave any physics-type of effects as highest as possible...
point is to burden CPU as much as possible with all the tasks it should be doing to the max, and on the other side not to bottleneck graphics card... going on VGA and SVGA resolutions isn't answer... as simple as that!
Answer is MultiGPU system with tuned low Shader effects, and upped high geometry complexity ( 'cos CPU feeds geometry) and texture quality (modern engines are streaming engines so PCI Express traffic, and implementation counts) in 1080 rez!
Last edited by Nedjo; 09-08-2009 at 02:46 PM.
Adobe is working on Flash Player support for 64-bit platforms as part of our ongoing commitment to the cross-platform compatibility of Flash Player. We expect to provide native support for 64-bit platforms in an upcoming release of Flash Player following the release of Flash Player 10.1.
All my respect onethreehill
Bookmarks