The problem of the performance of fresh video drivers for AGP video cards based on Radeon HD 46xx series chips

Here is a small life story - but at the same time it is a story about the “blue screen of death”.

Based on considerations of reasonable frugality, I still use a home computer purchased in the first (and not the second) half of the two thousandth, and therefore equipped with an AGP bus, and not a PCI-E bus. For the time being, I was satisfied with the NVidia GeForce 7600 GS vidyukha, all the more so as nothing had ever been done better on NVidia chips for plugging into AGPs: the eight-thousand series was already on PCI-E. However, in the middle of May, faced with the need for hardware acceleration of the video player, which, when playing a variety of high-quality H264-encoded 720p video, began to stutter unacceptably, I unwittingly remembered: let NVIDia CUDA technology on AGP be unavailable, but its ersatz alternative DXVA 1 is supported ( for example, in CoreAVC ) on all ATi Radeon HD chips - two thousand, three thousand, four thousand, five thousand and newer series.

It was then that I paid attention to the SAPPHIRE HD 4650 AGP vidyuhi, based on the eponymous Radeon chip. I was also pleased by the fact that (according to Wikipedia), NVidia GeForce vidyuhi seven thousand series were published in 2005 and 2006 and still supported only DirectX 9 with model shaders 3.0 - as for the more recent Radeon HD chips of four thousand series , they are shown there as published much later (in 2008) and supporting DirectX 10.1 and shader model 4.1.
')
After a little more thought, I decided to pay hundreds of rubles more to get a slightly more sophisticated chip of the same 4600th generation as part of the PowerColor HD4670 1GB DDR3 AGP product. The difference between 4650 and 4670 is not particularly great, but still the latter is a bit cooler.

_{I hardly punished myself with this steepness, because the PowerColor HD4670 1GB DDR3 only works on AGP 8x, and the SAPPHIRE HD 4650 also works on AGP 4x - respectively, if I remembered incorrectly that my home computer supports AGP 8x, then money would go down the drain (or, for example, it would be necessary to negotiate a money back). Fortunately, my computer is not old enough not to understand AGP 8x.}

Of course, the first thing after installing the video card should have thought about updating the drivers. At least because the hardware acceleration in Firefox 4 requires drivers of at least version 8.741 . Alas, the PowerColor website offers this version of the old drivers version 8.712, no more!

The logical way out of the impasse seemed to be the ability to download and install the latest AMD drivers. Unfortunately, this possibility turned out to be an exit to even greater horror and hopelessness, because the new AMD drivers, being installed on this vidyuhi, without delay cause BSoD (“blue screen of death”) when the system boots. It turns out that this problem (that is, the incompatibility of the AGP versions of Radeon with the new AMD drivers) exists for a hell of a lot (as early as last year), but no one was really honored to fix it. I myself could not even suspect such nasty things in advance, so that before I accidentally found out the story about it on the AMD user forum, I managed to try a little less than a dozen different versions of drivers that came out in the last six months, unsuccessfully, of course; and carried on without any reason until one in the morning. The situation is complicated by the fact that AMD does not provide AGP-versions of its former drivers (and only the AGP-version of the last driver, not a damn good one).

The desired driver (in October 2010) had to be searched in the archive on the Sapphire website : as you remember, Sapphire also produces AGP video cards based on Radeon chips. True, this archive turned out to be pervertedly organized: if a file does not appear on one of the proposed “mirrors”, then it is not proposed to choose another “mirror”; and since the rest of the page is already blocked by the “modal window” of the dialog, all that remains is to press the “Back” button and re-search for the desired, re-hammer the model of the vidyuhi and the operating system.

Having reached the desired result, first of all I tuned up the brightness and contrast, and then it was useful to compare the performance of the “new” video (PowerColor HD4670 1GB DDR3 AGP) with the old one (NVidia GeForce 7600 GS) using the browser performance test with HTML5 Canvas .

This test gave 11 fps or 12 fps (frames per second) in Firefox 4 on top of Direct3D 9 on top of NVidia GeForce 7600 GS. And now it has produced 16 fps or 17 fps in Firefox 4.0.1 over Direct3D 9 over PowerColor HD4670 1GB DDR3 AGP. Consequently, video performance has increased by ½ times, unless it is the effect of the difference between Firefox 4 and Firefox 4.0.1 (which I allow myself to doubt). The video player also began to use DXVA.

It would seem to me good. However, this story does not have a satisfactory happy ending. See for yourself how bad it looks. A large international corporation continues to provide on its website only those new versions of video drivers that immerse AGP variants in the “blue screen of death”, and this is only known informally. Only last year's drivers from alternative sources are suitable for use on AGP. At the same time, AMD is all the same, because AGP variants are produced by third-party firms in the third world, and officially, Radeons are released only in the form of PCI-E.

Can this go on forever?

No, of course not. Sooner or later, something new will appear in the new video drivers (for example, 3D video support on the Internet, which NVidia already has ), and then all the buyers of AGP video (and I, too) will be left behind for a bright future. The only alternative is to upgrade to PCI-E - the purchase of a new motherboard, a new processor, a new RAM, a new video card, a new cooler, finally.

Forced upgrade.