Finally. We're finally getting somewhere interesting in the graphics industry. Although they're sure to return, the days of reviewing $600 graphics card after $600 graphics card are on hiatus, and instead we're reviewing a new class of mainstream cards with earth-shattering performance.

NVIDIA's GeForce 8800 GT kicked off the trend, in one fell swoop making almost all of NVIDIA's product line obsolete thanks to the high performance and low price tag (we'll talk about that last part shortly). But what we saw there wasn't a fluke, it was a preemptive strike against AMD, who have been hard at work on an affordable GPU of their own.

This new product, like the 8800 GT, would be aimed squarely at the $150 - $250 market segment, something both AMD and NVIDIA did a horrible job at with mainstream releases earlier this year (2600 and 8600 both sucked guys).

Introducing the RV670

AMD's two new graphics cards launching today are both based off a new GPU, referred to internally as the RV670. The basic architecture of the hardware is largely unchanged from R600; there has been some additional functionality added, and a great deal of internal bandwidth removed, but other than that this is very much an R600 based part.

The biggest news of this part is that it is fabbed on a 55nm TSMC process. This is a half-node process based on 65nm technology, giving AMD an advantage in die size (cost) and potentially clock speed and/or power.

Historically, AMD's RV series has been a cost cut version of their R series designed for lower end volume parts, and that's where RV670 started. Right of the bat, half the external and internal memory bandwidth of R600 was cut out. External bandwidth dropped from 512-bit to 256-bit, but AMD stuck with 8 memory channels (each dropped from 64bit to 32bit).

Internally, the ring bus dropped from 1024-bit to 512-bit. This cut in bandwidth contributed to a significant drop in transistor count from R600's ~720M. RV670 is made up of 666M transistors, and this includes the addition of UVD hardware, some power saving features, the necessary additions for DX 10.1 and the normal performance tuning we would expect from another iteration of the architecture.

Processing power remains unchanged from the R600; the RV670 features 320 stream processors, 16 texture units and 16 redner back-ends. Clock speeds have gone up slightly and memory speeds have increased tremendously to make up for the narrower memory bus.

The RV670 GPU is also fully PCI Express 2.0 compliant like NVIDIA's G92, the heart and soul of the GeForce 8800 GT.

New Features you Say? UVD and DirectX 10.1
Comments Locked

117 Comments

View All Comments

  • ViRGE - Thursday, November 15, 2007 - link

    Testing
  • dustinfrazier - Thursday, November 15, 2007 - link

    Going on a year for Nvidia dominance and boy does it feel good. I bought my 8800gtx pair the first day they were available last year and never expected them to dominate this long. God I can't wait to see what comes out next for the enthusiasts. It get the feeling it is gonna rock! I really wanna see what both companies have up their sleeves as I am ready to retire my 8800s.

    I understand that these latest cards are great for the finances and good energy savers, but what does it matter if they already have a hard time keeping up with current next gen games at reasonable frame rates, 1920x1200 and above? What good does saving money do if all the games you purchase in 08 end up as nothing but a slide show? I guess I just want AMD to release a card that doesn't act like playing Crysis is equivalent to solving the meaning of life. Get on with it. The enthusiasts are ready to buy!
  • abhaxus - Thursday, November 15, 2007 - link

    err, tried to do a hardocp logo and it hid everything in the previous post.
    text is:

    I've read that... but then I've also read on AT and that with current bios releases the asus boards are fine to around 360-400 FSB. I haven't O/C'ed an intel chip since the Celeron 300A so I am pulling my hair out trying to decide if it's worth it to plan for going SLI or just get a P35 board and stay with a single card.

    sorry to go so OT. the article was very good in typical anand style.
  • JonathanYoung - Thursday, November 15, 2007 - link

    Just browsing through the article and this graphic caught my eye:

    Monitors command buffer to *ASSES* level of GPU utilization

    Not sure if this is an AMD or AT graphic, but you guys might want to correct it!
  • Anand Lal Shimpi - Thursday, November 15, 2007 - link

    That'd be an AMD graphic, if I had an editable source I'd correct it, but all I've got is the PDF :)

    Take care,
    Anand
  • imaheadcase - Thursday, November 15, 2007 - link

    I have heard quite a few people dislike those line graphs you use, the eyes just don't register lines well vs bars (one reason why long lines are not put in center of roads and they use bars to make the line) Why not stick to bar graphs like you do in the power consumption page?

    The eyes like things to conform to a shape, or should I say the brain. :) Quick glance at a bar graph is easier for the brain to compute than following lines.
  • strikeback03 - Thursday, November 15, 2007 - link

    Regarding lines on road, this is somewhere on US83 between La Pryor and Leakey in southern Texas.

    http://img.photobucket.com/albums/v315/strikeback0...">http://img.photobucket.com/albums/v315/strikeback0...

    I have no idea what that road marking means.

    As for the line charts, I like them better than a multiple bar chart would to display all the same info on a single chart.
  • Anand Lal Shimpi - Thursday, November 15, 2007 - link

    It's always tough finding a good balance, since I can cram so much more information into a line graph than a bar graph. I've just been toying with these things for the 8800 GT and this review, I'll see if I can come up with something better for the next round :)

    Take care,
    Anand
  • feraltoad - Thursday, November 15, 2007 - link

    Can't please everyone I guess. I really like the line graphs. I think it is much easier to compare cards scaling across resolutions and gives a better overview of performance in relation to one another.

    You could use hand puppets and then everyone would be happy. I know I would :)
  • JNo - Thursday, November 15, 2007 - link

    I second that - lines ftw

Log in

Don't have an account? Sign up now