Intel HD 4000 Explored

What makes Ivy Bridge different from your average tick in Intel's cycle is the improvement to the on-die GPU. Intel's HD 4000, the new high-end offering, is now equipped with 16 EUs up from 12 in Sandy Bridge (soon to be 40 in Haswell). Intel's HD 2500 is the replacement to the old HD 2000 and it retains the same number of EUs (6). Efficiency is up at the EU level as Ivy Bridge is able to dual-issue more instruction combinations than its predecessor. There are a number of other enhancements that we've already detailed in our architecture piece, but a quick summary is below:

— DirectX 11 Support
— More execution units (16 vs 12) for GT2 graphics (Intel HD 4000)
— 2x MADs per clock
— EU can now co-issue more operations
— GPU specific on-die L3 cache
— Faster QuickSync performance
— Lower power due to 22nm

Although OpenCL is supported by the HD 4000, Intel has not yet delivered an OpenCL ICD so we cannot test functionality and performance. Update: OpenCL is supported in the launch driver, we are looking into why OpenCL-Z thought otherwise. DirectX 11 is alive and well however:

Image quality is actually quite good, although there are a few areas where Intel falls behind the competition. I don't believe Ivy Bridge's GPU performance is high enough yet where we can start nitpicking image quality but Intel isn't too far away from being there.


Current state of AF in IVB

Anisotropic filtering quality is much improved compared to Sandy Bridge. There's a low precision issue in DirectX 9 currently which results in the imperfect image above, that has already been fixed in a later driver revision awaiting validation. The issue also doesn't exist under DX10/DX11.


IVB with improved DX9 AF driver

Game compatibility is also quite good, not perfect but still on the right path for Intel. It's also worth noting that Intel has been extremely responsive in finding and eliminating bugs whenever we pointed at them in their drivers. One problem Intel does currently struggle with is game developers specifically targeting Intel graphics and treating the GPU as a lower class citizen. I suspect this issue will eventually resolve itself as Intel works to improve the perception of its graphics in the market, but until then Intel will have to suffer a bit.

Discrete GPU Gaming Performance Intel HD 4000 Performance: Crysis Warhead
Comments Locked

173 Comments

View All Comments

  • ijozic - Thursday, April 26, 2012 - link

    Maybe because people who prefer to have the IPS screen would also like to have support for graphics switching to have a nice battery life while not doing anything GPU intensive. This was the one thing I expected from Ivy Bridge upgrade and NADA.
  • uibo - Monday, April 23, 2012 - link

    Does anyone know if the 24Hz issue has been resolved?
  • uibo - Monday, April 23, 2012 - link

    nevermind just saw the htpc perspective review
  • anirudhs - Monday, April 23, 2012 - link

    I didn't notice that issue. 23.976*1000 = 23976 frames, 24 * 1000 = 24000 frames, in 16 mins 40 secs. So that's about one second of mismatch for every 1000 seconds. I could not notice this discrepancy while playing a Blu Ray on my PC. Could you?
  • Old_Fogie_Late_Bloomer - Monday, April 23, 2012 - link

    Okay, well, I'm pretty sure that you would notice two seconds of discrepancy between audio and video after half an hour of viewing, or four seconds after an hour, or eight seconds by the end of a two-hour movie.

    However, the issue is actually more like having a duplicated frame every 40 seconds or so, causing a visible stutter, which seems like it would be really obnoxious if you started seeing it. I don't use the on-board SB video, so I can't speak to it, but clearly it is an issue for many people.
  • JarredWalton - Monday, April 23, 2012 - link

    I watch Hulu and Netflix streams on a regular basis. They do far more than "stutter" one frame out of every 960. And yet, I'm fine with their quality and so our millions of other viewers. I think the crowd that really gets irritated by the 23.976 FPS problems is diminishingly small. Losing A/V sync would be a horrible problem, but AFAIK that's not what happens so really it's just a little 0.04 second "hitch" every 40 seconds.
  • Old_Fogie_Late_Bloomer - Monday, April 23, 2012 - link

    Well, I can certainly appreciate that argument; I don't really use either of those services, but I know from experience they can be glitchy. On the other hand, if I'm watching a DVD (or <ahem> some other video file <ahem>) and it skips even a little bit, I know that I will notice it and usually it drives me nuts.

    I'm not saying that it's a good (or, for that matter, bad) thing that I react that way, and I know that most people would think that I was being overly sensitive (which is cool, I guess, but people ARE different from one another). The point is, if the movie stutters every 40 seconds, there are definitely people who will notice. They will especially notice if everything else about the viewing experience is great. And I think it's understandable if they are disappointed at a not insignificant flaw in what is otherwise a good product.

    Now, if my math is right, it sounds like they've really got the problem down to once every six-and-a-half minutes, rather than every 40 seconds. You know, for me, I could probably live with that in an HTPC. But I certainly wouldn't presume to speak for everyone.
  • anirudhs - Tuesday, April 24, 2012 - link

    I will get a discrete GPU and then do a comparison.
  • anirudhs - Monday, April 23, 2012 - link

    a discrete GPU! I could use a bump in transcoding performance for my ever-growing library of Blu-Rays.
  • chizow - Monday, April 23, 2012 - link

    Looks like my concerns a few years ago with Intel's decision to go on-package and eventually on-die GPU were well warranted.

    It seems as if Intel will be focusing much of the benefits from smaller process nodes toward improving GPU performance rather than CPU performance with that additional transistor budget and power saving.

    I guess we will have to wait for IVB-E before we get a real significant jump in performance in the CPU segment, but I'm really not that optimistic at this point.

Log in

Don't have an account? Sign up now