Can anyone show me test with 5850 CF vs single 480?
Printable View
Can anyone show me test with 5850 CF vs single 480?
@mmorpgfocus
Sapphire 5870 Toxic 2GB is 925MHz/5000MHz
http://www.hexus.net/content/item.php?item=23079
What was disturbing about anand's review is the lack of any mention of real competition. For instance, no mention of the 5970 being stiill faster, although obvious, but at the summary they mention the 480 being the fastest card (needs clarification - fastest single GPU). Also they left out the fact that with a small OC, the 5870 equals a stock 480, so that there is no clear winner here.
I felt like anand finally gave in to nVIDIA threatening to keep them out of the review circle if anand didn't produce a worthy review. Beauty is in the eyes of the beholder, I guess.
After reading the majority of these reviews I have decided to order a FC block for the 5870 and give it a proper OC. I game at 3840x1024 and from what I've seen, even at stock, the 5870 is very close in performance at this level to the 480. It even edges it out in some game tests. I think the FC water block will give me the edge in performance at a better price/performance ratio.
When I purchased my 5870 on release night it sold for $380. Add that to the cost of my new FC EK GPU block with backplate and we're up to 499. So really in terms of price, it's kind of a toss up at this point, but there is still value in having a very fast card (5870) for six months instead of waiting. Kudos to the sites that OC'd the 5870 and revealed that not only does it contend with the 480, in many ways it is still a better product. The 480 idles at 90C so for air that pretty much limits and real overclock. By the time one factors in water blocks, it seems safe to say that there will be a very high price premium ($650??) for the nVIDIA product. Feels like the 8800GTX launch all over again, minus the clear performance but keeping the high price, and adding much more heat.
Yeah, cause that would work with reviewers if it ever happened :rolleyes: Ridiculous.
I also think you're forgetting you can overclock any card.. why do you say there's no difference when you can overclock the GTX 480 as well? Then you're back to where you are now. :shrug:
ill refrain from trolling
but its hard :lol:
True that, I think I meant that since these new nvidia chips run so hot, I'm thinking the OC potential may not be the same. With water that can change, but now the price premium for nvidia really doesn't seem justified. I liked this interesting chart that pits mild air OC's against one another. If one were to invest in a $100 WB for the 5870, it seems like a better option than to go all out and purchase a 480. I wonder what a 1000/1250 ati5870 vs 800MHz nvidia 480 looks like?
Here's the chart I thought was pretty interesting. You can see the mildy oc'd MSI 5870 edges out the stock 480. Then notice MSI and the 480oc have a 100MHz core OC (the 480 edges out the MSI OC again), so if the 480 can't OC as far, the 5870 wins (for everyday, so LN2 could change that). From legit reviews.
http://www.legitreviews.com/images/r...rk_vantage.jpg
Already tried that, we tested using Furmark and GPUTool, I also suspected it may have been a power saving feature. nVidia's official confirmation on this is that there IS a bug and that since April 2009 their forceware drivers disable PCI-E Gen 2.0 on boards which do not have a BIOS with the fix.
A couple of months later Intel resolved the issue :)Quote:
Gen2 has been disabled on X38 for quite some time now, since around last year April to be exact. The reason behind this is due to a bug in X38 and the new SBIOS from Intel still shows problem. So our decision is to disable gen2 on X38
Testing has been done on 9800GT, 9800GTX+, 9600GT, GTX 280, GTX 285 all of these suffer from the Fallback to PCI-E 1.1 on several different ASUS boards.
However, when testing with a 9800GX2 and a GTX 295 the drivers reported PCI-E Gen 2.0, perhaps the NF200 was reporting PCI-E 2.0?
John
It idles at ~50 give or take, 100% load temp are around 92, but in the reviews where they had screens of precision like tool they show max fan cycle duty @ 55%, which is kinda low. I guess all they need to to is increase fan ramping delta for load temps to drop to 80ies.
Even at 92 load temps it's proves nothing, if they say that chip can handle it, then it's fine, lots of people use some personal factless logic to determine what temps are safe and what not.
Everybody seems to forget how that 5970 have almost the same load temps at stock cooler and it runs just fine.
Then ATI have a neverending issues with PWN that are hitting ridiculously 100+ temps and while 5850/5870 are not that much affected, on 5970 it can cause throttling. Nvidia's PWN design never had this issues, probably cause their coolers are better designed to cool them.
All i'm saying that there no clear winner, they are pretty much a valid competitors at this point, given that Fermi has only early beta divers to work with at this point while ATI had 6 month work put in the drivers, picture might very well change in a couple of driver revisions for Fermi. I also have a feeling that heavy DX11 titles might run better on Fermi due to their architectural focus on tessellation among other things.
because they are cheap
First review pitted against the 2GB 5870
http://www.hexus.net/content/item.php?item=24024&page=6
http://www.hexus.net/content/item.php?item=24024&page=7
http://www.hexus.net/content/item.php?item=24024&page=9
http://www.hexus.net/content/item.ph...=24024&page=11
Even if you look at the last page:
http://www.hexus.net/content/item.ph...=24024&page=12
All of the games they showed @ 2560x1600 w/ 8xAA are playable on a 5850. Hexus is not doing a good job making the case for the gtx480 (conclusion reeks of nvidia pee are).
I find this very weird
http://img.inpai.com.cn/article/2010...0dfc06ba10.png
http://img.inpai.com.cn/article/2010...f7532f48cf.png
http://en.hardspell.com/doc/enshowco...23&pageid=6891
So with increase in resolution the ATI's card performance actually increases instead of decreasing??