Morning, all. I hope you had a good Thanksgiving and holiday weekend. Mine was restful and low-key, which is what I was craving after an exceptionally busy past few months, with lots of travel interspersed with major GPU reviews. This year, as often happens after a busy season of new releases, I've been in some pain in my back, neck, and shoulder from too much working at a desk. I was able to use the break to recover some.
With that said, things have been busy in Damage Labs lately, even though I haven't published much. I've run into some technical snags of various sorts. One of the first things I sought to do after the GTX 780 Ti review was to measure the performance of AMD's new CrossFire mechanism in the Radeon R9 290 and 290X. My plan was to test multi-GPU performance on a 4K monitor using FCAT, but I ran into an odd problem with color consistency in video capture of the FCAT overlay with the R9 290X. That turned out to be a show-stopper; I couldn't get good performance results. I just had to shift my emphasis and hope that Nvidia, who developed FCAT, could fix the problem. So far, I'm still waiting.
I figured, in the interim, I could do an article on 4K performance with the latest high-end graphics cards using only Fraps, which is generally sufficient for single-GPU testing. To make that happen, I did a fresh install of Windows 8.1 on my GPU test rigs and installed a number of the latest games, including BF4, Arkham Origins, CoD: Ghosts, and AC4: Black Flag. I twiddled with each game, figuring out a fitting test scenario and the appropriate image quality settings for 4K testing. I got well down the road on that project, nearly ready to publish, when Nvidia uncorked a driver claiming "up to 50%" performance increases in a number of games I'd tested. And my results were no longer current.
At that point, rather than immediately re-testing the GeForce cards, I turned my attention yet another project in the same realm, and it turned out to be an awful lot of work. You should see the results of that effort soon. Once that's in the can, I'll probably get back to the 4K testing in some form. I should be primed to produce some interesting results in short order, since I've already put in a lot of the work.
I'll try to write up some general impressions of 4K gaming along the way, too. The higher resolution is easy to appreciate in some cases, but I think fluid animation at current resolutions is probably more important than adding more detail onscreen. Also, I think perhaps texture filtering routines and shader sampling standards need to be tweaked to account for the additional sharpness of high-PPI displays. It's weird to say, but with this much fidelity, I'm back to wanting supersampling more than ever.
|Gigabyte SA-SBCAP3350 puts formidable power on a single board||13|
|Alphacool Eisblock HDX-2 and HDX-3 help M.2 SSDs beat the heat||11|
|Corsair Lighting Pro Expansion Kit lets builders turn up the lights||10|
|Adata D16750 power bank is tougher than the average juice pack||16|
|Deals of the week: fast memory, an AM4 motherboard, and more||18|
|Corsair RMx White Series PSUs take a walk on the snowy side||24|
|Intel crams 100 GFLOPS of neural-net inferencing onto a USB stick||42|
|Toshiba's XG5 1TB NVMe SSD reviewed||10|
|Microsoft and Johnson Controls put Cortana in a thermostat||26|