ATI HD 5870 Scaling Performance: X58 vs P55 Showdown
by Gary Key on September 29, 2009 12:00 PM EST- Posted in
- Motherboards
FarCry 2
We utilize the Ranch Small demo file along with the Playback Action demo to see the differences between GPU and CPU centric benchmarks. The Ranch demo is GPU centric while the Playback demo tends to be CPU centric. We run each benchmark five times and report the median score.
First off, FarCry 2 has always performed extremely well on the X58 chipset compared to other chipsets with the ATI video cards. This game responds very well to PCIe bandwidth, something the X58 has in droves. The stock HD 5870 single card x8 configuration is 6% slower in average frame rates and 10% slower in minimum frame rates in this particular benchmark. Even the 4.2GHz single card result is slower than either the stock P55 or X58 setups.
The stock single card HD 5870 P55 platform is 3% slower than the X58 even with a 100MHz processor advantage in turbo mode. At 4.2GHz, both platforms are about even although minimum frame rates are about 4% better on X58. However, even with this benchmark advantage, there was no difference in actual game play, especially considering minimum frame rates are above 70fps in each instance.
The P55 results improve slightly as we move to a more CPU centric benchmark. The stock HD 5870 single card x8 configuration is only 2% slower in average frame rates and only 1% in the overclocked results. Comparing single card results between the P55 and X58, we see a 2% difference in favor of the X58 at stock speeds while at 4.2GHz the P55 finishes slightly ahead of the X58.
The pertinent data for CrossFire scaling is in the tables below. What we are looking for is the percentage speedup going from one to two HD 5870s on X58 and P55. In theory, X58 should have improved percentages because each GPU gets 16 PCIe 2.0 lanes while Lynnfield only provides 8 PCIe 2.0 lanes per GPU.
FarCry 2 CrossFire Scaling – Average Frame Rates
| ATI HD 5870 CF Scaling | FarCry 2 – Ranch Small | FarCry 2 – Playback Action | FarCry 2 – Ranch Small 4.2GHz | FarCry 2 – Playback 4.2GHz |
| Intel Core i7 920 (X58) | 66.8% | 9% | 81.8% | 42.5% |
| Intel Core i7 860 (P55) | 59.8% | 10.6% | 75.8% | 44.4% |
Based on our single card results, there are not a lot of surprises here. At stock speeds, the X58 has a 7% scaling advantage over the P55 and 6% when overclocked in the GPU centric Ranch demo. In the Playback Action benchmark, the results favor the P55 by almost 2%.
FarCry 2 CrossFire Scaling – Minimum Frame Rates
| ATI HD 5870 CF Scaling | FarCry 2 – Ranch Small | FarCry 2 – Playback Action | FarCry 2 – Ranch Small 4.2GHz | FarCry 2 – Playback 4.2GHz |
| Intel Core i7 920 (X58) | 41.1% | 3.6% | 77.9% | 30.5% |
| Intel Core i7 860 (P55) | 32.4% | 3.6% | 78.4% | 28.3% |
Minimum frame rates and scaling heavily favor the X58 in our stock clock speed results using the Ranch demo. Although frame rates still favor the X58 in this demo when overclocked, the scaling on the P55 is slightly better. The stock results in the Playback Action demo are a dead heat with a 2% advantage to the X58 when overclocked.
When it came to actual game play, there were no differences between either platform in the game. In fact, it was very difficult to discern which system was being utilized. The key giveaway was the foot warming heat coming from our case with the X58 overclocked. Our ambient temperature in the test room rose 2.1C over the course of testing with the X58 installed compared to 0.7C with the P55.
Facebook
Twitter
RSS
85 Comments
View All Comments
tamalero - Thursday, October 1, 2009 - link
and whats your opinion about SiliconDoc Gary?it seems what the Zorro is for ATI, the SiliconDoc is for Nvidia (but weirder and more annoying)
Gary Key - Friday, October 2, 2009 - link
Honestly, I fully believe in trying to adhere to a no-ban policy as long as the discussions are constructive and do not turn into personal attacks. That said, SiliconDoc should be banned ASAP based on his "useless" spamming and he probably will be very shortly. His actions have ruined the last two articles.Our favorite the zorro, thezorro, aka SnakeOil has and will continue to be banned if his actions continue. Eventually he will run out of IP addresses or will control the Internet. :)
TA152H is a different animal, instead of simple shock jock posting, he tends to get extremely personal in his attacks not only against me, but also Anand and now other readers. He has not done this at other sites, which makes me believe I ran over his dog and Anand took out his cat recently. He is on a very short leash.
If I screw up and get called out on it, fine, I deserve it and the comments section is a way for people to communicate it. We do care what is said and pay attention in order to improve our articles. If you have a personal agenda with me, well, use the email or PM system and we will try to work it out.
However, considering the fact that our testing methodology is extremely sound and we play by the book, I find his comments to be perplexing as he wants us to change system settings for one platform to make it look better but doing the same tuning on the other platform is forbidden. At first I thought he was just trying to drum up controversy or look good at Toms, until I spoke with the editors at Toms and lets just say, that was a conversation I cannot repeat in public. ;)
Scheme - Friday, October 2, 2009 - link
If people want to troll they'll troll, no matter how many times you ban an IP they've used to post.Maybe you should consider making the registration process more rigorous, perhaps have it use forum accounts, and start moderating comments. More effort required to circumvent a ban, combined with less reward if your posts quickly get deleted may mean less incentive to troll.
At the moment it feels like it's open season for a small number of idiots to ruin the comments sections.
strikeback03 - Thursday, October 1, 2009 - link
Will TA152H choose another failed German fighter aircraft to name himself after if banned?mesiah - Thursday, October 1, 2009 - link
Gary, I don't know how you guys can read some of the comments and not want to just flip out sometimes. Maybe what you should do to make things a little more entertaining is this... Just like clubs have muscle around to work over annoying guys and toss them out on the street, you should hire a full time bully to verbally harass trolls before you ban them. It wouldn't really fix anything, but it would make my day more entertaining :)james jwb - Wednesday, September 30, 2009 - link
Anandtech is a massive, respected technical site whose audience has always been well-rounded, unlike say, Engadget, who rely on fanboy comments to fuel large debate. This site has never been like this, it's not a core of anything around here.Rudeness should not be tolerated, and ta152h is a piece of work imo. He deserves a ban on multiple levels. This guy has some issue with this site, it's that simple.
Maybe toleration of trolls like this can be left to their demise once you have a comment system where the audience can vote down idiotic comments, like engadget or dailytech.
adam92682 - Wednesday, September 30, 2009 - link
How about testing a p55 system with the NF200 on it?TA152H - Tuesday, September 29, 2009 - link
Another bad review by Gary.I was hoping you'd have learned something about testing by now, but, apparently not.
For the rest of you, you'll notice Gary is up to his old tricks. He clocks the Lynnfield slightly higher, and, also, clocks the uncore considerably higher.
Keeping in mind this increases memory access, as well as L3 cache speed, this is a meaningful increase. Also keep in mind that the cores use this uncore to communicate with each other. So, the uncore clock speed isn't immaterial.
On the plus side, he did mention the fact the Lynnfield needs higher voltages to clock the same as the Bloomfield. Quite a bit higher. Makes you wonder why he never mentions maximum overclocks, since they would seem to favor Bloomfield.
This type of bad testing is shameful, for anyone who has any sense of shame. Even with his clear bias for the brain-damaged (yes, I know at least a few of you were waiting for me to say it) Lynnfield, with a higher clock speed (albeit slightly), and higher uncore, he still couldn't get it to go faster.
Don't waste your money on Lynnfield for gaming. In real world, where you weren't clocking the uncore much faster, the Bloomfield would win by more, can overclock higher, and has a six core coming out, compared to a two core. Yes, the six core will probably be too expensive, for now, but it's at least an upgrade, and will be cheaper in the future. It's hard to see moving from quad core to dual core as any upgrade.
By the way, is anyone else wondering what Intel will actually be making on 32nm? With only the Gulftown for x58, and only a dual core for P55, what are they going to do with all their 32nm capacity? The Gulftown is big, but shouldn't sell that great, the dual core with GPU should sell very, very well to the masses, but it's small, especially since it's only the processor that's 32nm. There's got to be something else that will be coming out we haven't heard of. There's too much capacity, and too few products.
Intel is doing a really bad job of things lately. They release a lobotomized processor that isn't segmented properly, still fail to release a decent chipset for Atom, don't announce a single quad core 32nm processor, and still haven't lowered the power use on the x58 chipset, which has been out for a year. They're so uneven.
PorscheRacer - Saturday, October 10, 2009 - link
I'm guessing Larrabee would be using 32nm production also...goinginstyle - Tuesday, September 29, 2009 - link
You have to be really brain damaged TA152H."For the rest of you, you'll notice Gary is up to his old tricks. He clocks the Lynnfield slightly higher, and, also, clocks the uncore considerably higher. "
Looking at the screenshots both of the boards are at their stock ratios with their Bclk at 133.5, the 920 hits a 21x multi with turbo enabled and the 860 hits a 22x multi with turbo enabled, as designed and implemented by Intel. Or is that a concept that is above your head?
When overclocked, the p55 bclk is at 200.7 and the x58 is at 200.4. That is due to the clock generator. It would be impossible for them to change that and the resulting 5Mhz advantage does nothing for any score at 4.2GHZ. The uncores are set to stock ratios based on the screen shots. This means he did not tamper with the settings.
If you actually owned either system you would know a 200MHz uncore clock improvement does nothing for the benchmarks. Once again, you are grasping at straws here trying to defend a platform that is the electric companies best friend.
Where is that review of yours showing how any of your comments have validity? All you do is trash every article here and I hope they just ban you tonight for obvious flame baiting. None of your comments are true dude. So go back to doing your cut and paste articles over at Toms.