Dear forum members,
We are delighted to inform you that our forums are back online. Every single topic and post are now on their places and everything works just as before, only better. Welcome back!


Discussion on Article:
DirectX 11 in the Open: ATI Radeon HD 5870 Review

Started by: hiVe | Date 09/22/09 11:47:04 PM
Comments: 73 | Last Comment:  09/02/16 03:59:40 AM

Expand all threads | Collapse all threads


Please fix the following sentence on "Page 2":
"Then, the Radeon HD 5870 is the same as the Radeon HD 4870 in terms of potential computing power as well as texture and raster processors"
"Then, the Radeon HD 5870 is the same as the Radeon HD 4870 X2 in terms of potential computing power as well as texture and raster processors"

0 0 [Posted by: hiVe  | Date: 09/22/09 11:47:04 PM]
- collapse thread

Will be corrected ASAP, thanks!
0 0 [Posted by: Vader@Xbit  | Date: 09/23/09 06:21:11 AM]

Power consumption measurements seem whacked. Likely due to a severe CPU bottleneck limiting GPU load. Seriously, why bother running such a card with Prescott Pentium 4 3.6GHz. This also applies to other xbitlabs reviews. Other sites that measured GPU power consumption (card only) report ~150W peak in 3DMark.
0 0 [Posted by: largon  | Date: 09/23/09 03:09:39 AM]
- collapse thread

Read carefully:
FurMark provides a different picture. It makes full use of all GPU resources and the new card’s peak power consumption is as high as 160 watts.
0 0 [Posted by: Vader@Xbit  | Date: 09/23/09 06:17:24 AM]

I remember someone (though I forget the name) talking about a plan to upgrade the testing rig for power consumption. Is that still in the works?
0 0 [Posted by: Joe Public  | Date: 09/23/09 11:50:54 PM]

I wish you had also included 2x 4890 CF in the test because in my country the HD5870 have exactly the same price as 2x HD4890.
0 0 [Posted by: zenmyo  | Date: 09/23/09 05:14:25 AM]

Ati second revolution with 5 type shader.Temperature compensator for third generation of Gddr5 memory,for Nvidia and old Ati card look like UFO .
0 0 [Posted by: Blackcode  | Date: 09/23/09 12:58:50 PM]

Good preview! The performance in NFS:Shift is outstanding. Maybe a little more discussion about this matter?
0 0 [Posted by: TAViX  | Date: 09/23/09 07:36:57 PM]
- collapse thread

It is because the game does not scale in SLI/CF.
0 0 [Posted by: jonup  | Date: 09/28/09 03:57:41 AM]

I would like to suggest the GPU reviewers at Xbitlabs to stop using the built in benchmarks and start using Fraps with recorded tracks from the games to measure the
average and minimum performance numbers.

I believe that the built in bemchmarks are being badly utilized by the drivers to get more performance numbers out of them that do not represent the actual performance numbers of the GPU.
0 0 [Posted by: Vlad  | Date: 09/24/09 09:52:24 AM]

Alexey and Yaroslav,

Do you think that Nvidia's Transparency Multisampling AA provides equal image quality to ATI's Quality Adaptive AA, let alone Performance Adaptive AA?
0 0 [Posted by: Bo_Fox  | Date: 09/30/09 06:56:29 PM]
- collapse thread

In our opinion they are very similar:
0 0 [Posted by: Vader@Xbit  | Date: 10/01/09 05:17:19 PM]
Thanks, but on that page, Fallout 3 already appears to be using alpha-AA by default for ATI cards when AA is being used, unlike with Nvidia cards.

Even Nvidia's best TR-SSAA mode does not clearly show the tree above the building in the lower right corner (the tree is almost non-existent), while all of ATI's modes show that tree so conspicuously and anti-aliased.

Also, you seem to base your conclusion on only one game, Half Life 2 Episode 2, in that since there was not a difference between ATI's Performance and Quality Adaptive AA, Nvidia's TR-MSAA is roughly equal to ATI's Quality Ad-AA. After closely looking at Nvidia's TR-MSAA, I see so many sampling errors on the fence that it is almost worse than leaving them with jaggies. It certainly seems that ATI's Ad-AA is at least equal to Nvidia's TR-SSAA, not TR-MSAA.
0 0 [Posted by: Bo_Fox  | Date: 10/11/09 01:35:34 AM]


Back to the Article

Add your Comment