Under the Hood of G84

So the quick and dirty summary of the changes is that the G84 is a reduced width G80 with a higher proportion of texture to shader hardware and a reworked PureVideo processing engine (dubbed VP2 as opposed to G80's VP1). Because there are fewer ROPs, fill rate and antialiasing capabilities will be reduced from the G80 as well. This isn't as necessary on a budget card where shader power won't be able to keep up with huge resolutions either.

We expect the target audience of the 8600 series to be running 1280x1024 resolution panels. Of course, some people will be running larger panels and we will test some higher resolutions to see what kind of capabilities the hardware has, but above 1600x1200 tests are somewhat academic. As 1080p TVs become more popular in the coming years, however, we may start putting pressure on graphics makers to target 1920x1200 as their standard resolution for mainstream parts even if average computer monitor sizes weigh in with fewer pixels.

In order to achieve playable performance at 1280x1024 with good quality settings, NVIDIA has gone with 32 shaders, 16 texture address units, and 8 ROPs. Here's the full breakdown:

GeForce 8600/8500 Hardware
GeForce 8600 GTS GeForce 8600 GT GeForce 8500
Stream Processors 32 32 16
Texture Address / Filtering 16/16 16/16 8/8
ROPs 8 8 4
Core Clock 675 MHz 540 MHz 450 MHz
Shader Clock 1.45 GHz 1.19 GHz 900 MHz
Memory Clock (Data Rate) 2 GHz 1.4 GHz 800 MHz
Memory Bus Width 128-bit 128-bit 128-bit
Frame Buffer 256 MB 256 MB 256MB / 512MB
Outputs 2x dual-link DVI 2x dual-link DVI ?
Transistor count 289 M 289 M ?
Price $200 - $230 $150 - $160 $90 - $130

We'll tackle the 8500 in more depth when we have hardware. For now, we'll include the data as reference. As for the 8600, right out of the gate, 32 SPs mean one third the clock for clock shader power of the 8800 GTS. At the same time, NVIDIA has increased the ratio of Texture address units to SPs from 1:4 to 1:2. We also see a 1:1 ratio of texture address and filter units. These changes prompted NVIDIA to further optimize their scheduling algorithms.

The combination of greater resource availability and improved scheduling allow for increased efficiency. In other words, clock for clock, G84 SPs are more efficient than G80 SPs. This makes it harder to compare performance based on specifications. Apparently stencil culling performance has also been improved, which should help boost algorithms like the Doom 3 engine's shadowing technique. NVIDIA didn't give us any detail on how stencil culling performance was improved, but indicated that this, among other things, was also tweaked with the new hardware.

Top this off with the fact that G84 has also been enhanced for higher clock speeds than G80 and we can expect much more work to be done by each SP per second than on 8800 hardware. Exactly how much is something we don't have an easy way of measuring as changes in efficiency will vary by the algorithms running on the hardware as well.

With 256 MB of memory on a 128-bit bus, we can expect a little more memory pressure than on the 8800 series. The 2 x 64-bit wide channels provide 40% of the bus width of an 8800 GTS. This isn't as cut down as the number of SPs; remember that the texture address units have only been reduced from 24 on the 8800 GTS to 16 on the 8600 series. Certainly the reduction of 20 ROPs to 8 will help cut down on memory traffic, but that extra texturing power won't be insignificant. While we don't have quantitative measurements, our impression is that memory bandwidth is more important in NVIDIA's more finely grained unified architecture than it was with the GeForce 7 series pipelined architecture. Sticking with a 128-bit memory interface for their mainstream part might work this time around, but depending on what we see from game developers over the next six months, this could easily change in the near future.

Let's round out our architectural discussion with a nice block diagram for the 8600 series:



We can see very clearly that this is a cut down G80. As we have discussed, many of these blocks have been tweaked and enhanced to provide more efficient processing. The fundamental function of each block remains the same, and the inside of each SP remains unchanged as well. The features supported are also the same as G80. For 8500 hardware, based on G86, we drop down from two blocks of Shaders and ROPs to one each.

Two full dual-link DVI ports on a $150 card is a very nice addition. With the move from analog to digital displays, seeing a reduction in maximum resolution on budget parts because of single-link bandwidth limitations, while not devastating, isn't desirable. There are tradeoffs in moving from analog to digital display hardware, and now an additional issue has a resolution. Now we just need to see display makers crank up pixel density and improve color space without reducing response time and this old Sony GDM-F520 can finally rest in peace.

In the video output front, G84 makes a major improvement over all other graphics cards on the market: G84 based hardware supporting HDCP will be capable of HDCP over dual-link connections. This is a major feature, as a handful of larger widescreen monitors like Dell's 30" only support 1920x1080 with a dual-link connection. Unless both links are protected with HDCP, software players will refuse to play AACS protected HD content. NVIDIA has found a way around the problem by using one key ROM but sending the key over both links. The monitor is able to handle HDCP connections on both links, and is able to display the video properly at the right resolution.

As for manufacturing, the G84 is still an 80 nm part. While G80 is impressively huge at 681M transistors, G84 is "only" 289M transistors. This puts it at nearly the same transistor count as G71 (7900 GTX). While performance of the 8600 series doesn't quite compare to the 7900 GTX, the 80 nm process makes smaller die sizes (and lower prices) possible.

In addition to all this, PureVideo has received a significant boost this time around.

Index The New Face of PureVideo HD
POST A COMMENT

60 Comments

View All Comments

  • poohbear - Tuesday, April 17, 2007 - link

    sweet review on new tech! thanks for the bar graphs this time! good to know my 512mb x1900xtx still kicks mainstream butt.:) Reply
  • tuteja1986 - Tuesday, April 17, 2007 - link

    Total disapointment :( ... Could even beat up a X1950pro. They really need to sell at $150 otherwise you would be better off buying a X1950GT or 7900GS for $150 to $160. At the current $200 to $230 price you could get a X1950XT 256MB which could destroy it but that GPU needs a good powersupply. Only thing going for a 7600GT is the DX10 support and Full H.264 , VLC , Mepg 4 support but that can be found on even other cards. Reply
  • Staples - Tuesday, April 17, 2007 - link

    I have been waiting several months for these cards and boy and I disappointed. I figured this month I would get a new PC since April 22 the prices of C2D also drop. My idea was to get a C2D600 and an 8600GTS but after their lack luster performance, my only option is an 8800GTS which is $50+ more. Not a huge difference but I am very compelled to wait until the refresh comes out and then maybe I can get a better deal. I really hate this senario where ATI is down, AMD is down and no competition is leading to high prices and crappy performance.

    Hopefully in another 6 months, AMD will be up to par on both their processors and CPUs. I will be holding on to my current system until then. I found it disappointing that these cards do not come with 512MB of memory but their performance is actually even more disappointing.
    Reply
  • JarredWalton - Tuesday, April 17, 2007 - link

    Well, the R6xx stuff from AMD should be out soon, so that's going to be the real determining factor. Hopefully the drivers do well (in Vista as well as XP), and as the conclusion states NVIDIA has certainly left the door open for AMD to take the lead. Preliminary reports surfacing around the 'net show that R600 looks very promising on the high end, and features and specs on the midrange parts look promising as well. GDDR4 could offer more bandwidth making the 128-bit bus feasible on the upper midrange parts as well. Should be interesting, so let's see if AMD can capitalize on NVIDIA's current shortcomings.... Reply
  • PrinceGaz - Tuesday, April 17, 2007 - link

    It might be a good idea to read more reviews before writing off the 8600 range.

    Over at , they found the 8600GTS easily beats the X1950Pro and even though the models they tested were factory overclocked, one of them had only a 5% core overclock and no memory overclock and was still well ahead of the X1950Pro. At worst the two cards were roughly even but in many tests the 8600GTS (with just 5% core o/c) was considerably faster. As say in the article link
    quote:

    New GeForce 8600 GTS Shines
    If you have been waiting for a DX10 video card capable of playing today's games, your wait is over. The GeForce 8600 GTS series GPU destroys ATI's current X1950 series for right around $220.


    So who do you believe? I guess I'll need to read several more reviews to see what's really going on.
    Reply
  • Spanki - Tuesday, April 17, 2007 - link

    I agree that that review it paints a different picture, but I have no reason to disbelieve it. I mean, they do their testing differently (try to find the best playable settings for each card), but it is what it is... I mean, the tables show the differences in the cards, at those (varrying settings) and so they therefore draw the conclusions they draw, based on that (with card X, I can enable option Y at these frame rates, but not on card Z - at least at the same resolutions). It's not like they tried to hide the settings they used or the frame-rates they got with each set-up. I found it an interesting perspective. ~shrug~

    Anyway, my personal opinion is that they neutered this chipset too much. There looks to be a substantial gap between 7900/8600 and 8800 level performance and the sweet-spot for this price-point would have been right in the middle of that gap... maybe they're planning a 8700 series chipset?
    Reply
  • Griswold - Tuesday, April 17, 2007 - link

    Thats a funny review. I'll stick to the other 90% that say this fish smells. Reply
  • GoatMonkey - Tuesday, April 17, 2007 - link

    Reply
  • yacoub - Tuesday, April 17, 2007 - link

    HardOCP's review disagrees with almost everyone else's results and also reads like a marketing advertisement for the product. I wouldn't give their review the time of day. Reply
  • PrinceGaz - Wednesday, April 18, 2007 - link

    They're normally very good which after reading HardOCP's review immediately after AT (whose I read first), I thought I should mention that not everyone found the 8600GTS to be slower than the X1950Pro.

    However, after reading several more reviews on other usually reliable websites, the consensus seems to be that the 8600GTS is well behind the X1950Pro, which does make HardOCP's finding seem very odd.

    I get the feeling that we're going to have to wait until nVidia get their drivers for this card sorted out as I suspect they are not all they could be, which they will hopefully have done by the time the HD2xxx series are launched, then the 8600/8500 cards can be retested and compared with their true competition.
    Reply

Log in

Don't have an account? Sign up now