AM3 Reference Diagram Disclosed

psyph3r writes "Chilehardware has released what appears to be a confidential image showing the future customer desktop AM3 reference boards for AMD and ATI. Here is an English site talking about this reference design image and the features it enables. 'The biggest improvement for this generation of chipsets is the audio and video capabilities integrated into the motherboard. The new features packed into these chipsets are beginning to look like standalone platforms. The RS780 supports DirectX 10 and has a UVD, which is similar to most High-end cards of today.'"

Comments

In other words, integrated

[Applekid]

Hasn't integrated audio and video been around forever?

Supporting DirectX 10 and all that is great and all, but, how fast will it be? I remember getting an nForce 4 integrated video board for my folks some time ago and it supported the latest DirectX versions and, while it ran all the nVidia eyecandy demos, it sure was slow.

I mean, TFA makes reference to Hypertransport 3.0 and all, but memory bandwidth is only part of pretty pixels.

Re:In other words, integrated

[Aladrin]

I don't understand the assumption that they can't embed a high-end graphics card in a motherboard? It hasn't typically been done since the resulting board would be seem to be insanely expensive until you realize that most of the cost is the graphics... But that doesn't mean they can't or won't.

Also, when you say 'decent graphics performance' ... I assume you're talking about games, since the Intel GMA 3xxx series does perfectly well with Compiz. It's every bit as fast as the ATI card in my work machine.

Re:In other words, integrated

[Svartalf]

What makes an IGP slower is a couple of things.

In the case of Intel, it's the memory bandwidth coupled with a distinct lack of Vertex Shader support.

In the case of AMD, it's the memory bandwidth coupled with a dramatically reduced/nonexistent support for Vertex Shaders.

In the case of NVidia, it's the memory bandwidth.

In the case with many IGPs, the combination of having to share RAM with the machine on it's own bus, along with no Transform, Clipping, and Lighting hardware acceleration (Little to no Vertex Shader hardware...) means for a very slow GPU overall. Now, having said this, the Hypertransport 3.0 interface may help on the bus speed, and if you're looking at the Unified Shader requirements for DX10, you might find that this may be a little better performer. It's not going to be like a PCI-E add-in card, but it may be serviceable for light to medium duty 3D stuff by itself because of those two things.

Re:In other words, integrated

[moderatorrater]

I don't understand the assumption that they can't embed a high-end graphics card in a motherboard I assume that it can be done but it hasn't. The reason I assume it hasn't been done is that the resulting motherboard would then have to go through testing and the video card would have to get integrated. In the meanwhile, the GPU market is moving forward and releasing new cards and marking down existing ones. In addition, the video card industry moves faster than the motherboard industry generally speaking, so while a good motherboard is useful 6 months later at roughly the same cost, a video card isn't. Most motherboard manufacturers would rather let enthusiasts with higher graphics requirements purchase the card separately and embed low-quality GPUs for people who don't need a better one.

Re:In other words, integrated

[Aladrin]

I agree totally... Except that conditions are a bit different. The improvements in graphics aren't as mind-blowing as they were 3+ years ago. I'm a graphics whore. I admit it. I'm one of those fools that buys $500 video cards.

But since the 7800s, things haven't been so urgent. The new cards aren't -that- much better than the old ones because games aren't pushing the limits as much. It used to be there were several games a year that required rigs that were insane. Now there's maybe 1 or 2. Most of the really fun games, the ones that focused on gameplay instead of maxing out a high-end system, just don't need that kind of power.

I keep looking at new cards, but really the only thing that draws me in is the 'silent' ones... I'm about sick of the noise. The extra power will be nice, but not necessary at all.

Re:In other words, integrated

[Chosen Reject]

It used to be there were several games a year that required rigs that were insane. Now there's maybe 1 or 2.

I haven't noticed that. What I've seen is that usually games will reach a plateau for a while (maybe a plateau with a slight incline) and then suddenly jump forward. For example, Doom 3 and Half Life 2 were both released withing a few months of each other, and Unreal Tournament 2004 was released only a few months before Doom 3 (UT2k4 required enough of a boost over UT2k3 that I included it). But then there wasn't much after that for a while. But this year we have BioShock, Crysis, and UT3, which all go towards upping the requirements quite a bit. Granted that is for all the eye-candy they can offer (which is what a $500 video card is for).

However, there has been a general trend in computing power in the last few years where people aren't upgrading that often. Some attribute it to computers being fast enough to run just about everything. Personally I think it's because we've been with Windows XP for so long, which is an argument for a another time. However, regardless of the reasons why people aren't upgrading that often, game developers are realizing that a lot of their money is coming from people still running DirectX 8 video cards and they had better make their games at least playable to them. Another reason is that the costs of producing a game are soaring, and developers also realize they cannot entirely rely on the hardware-upgraders for sales like they could in ye olden days of 10 years ago. So they let the market grow by allowing people with non-top-of-the-line computers to run their games. It's not a large expansion, but it is a growth nonetheless.

Nevertheless, people willing to put out "insane" amount of money for a good mobo and GPU, are not typically going to be the kind of people that would pay that same amount for the two to be one unit.

Re:Why usb 1.1 and 2.0? and why not use HT for the

[PetiePooo]

12 USB2.0 should be plenty for all your USB2.0 peripherals. I imagine the slower USB1.1 ports are a freebie in case you have USB1.1 devices that don't auto-negotiate well on a USB2.0 port... I wouldn't be surprised if most integrators don't even provide the pinouts to use them.

As for PCIe vs. HT, they're probably so similar in latency and throughput at that level that its just a difference in transistor count or something similarly insignificant.

Wasn't this "Confidential"?

[dilute]

OK, flame me and mod me -1, but if the Slashdot editors had good reason to believe this was actually confidential (and based on the translation of this article, this pretty plainly appears to be the case), and an unauthorized disclosure, why the editors here decide to carry the story? If someone submitted a story that said, "Here are documents I STOLE from Microsoft by breaking into the building" would Slashdot carry that? Where do you draw the line? Why does AMD's stuff have to be outed like this as a consequence of someone violating their confidence? Or maybe it's a deliberate leak (???)

Re:Wasn't this "Confidential"?

[A beautiful mind]

"Here are documents I STOLE from Microsoft by breaking into the building" would Slashdot carry that?

You were doing quite well until you wrote this sentence, but c'mon this is slashdot! Did you really have to ask a question like that?

Re:Wasn't this "Confidential"?

[rilister]

Does this:
http://slashdot.org/articles/04/02/12/2114228.shtml
answer your question?
(Slashdot's most visited story of all time, btw)

What I want from a motherboard...

[unfunk]

* No integrated Audio
* No Integrated Video

Is that really so hard? Integrated video is easy enough to avoid, but you just can't get a motherboard these days that doesn't have onboard audio. I'm sick of having to disable it whenever I get a new board, and the amount of space the jacks take up on the rear panel could be better used for more USB or Firewire ports.
I use an old Soundblaster Audigy for my sound needs, and it does everything I need. In hardware. Every time I buy a new motherboard, I test the onboard audio first, just to see if it's gotten any better than I last tried it.
So far, this card's lasted me four complete system overhauls, and at this rate, will last until a version of Windows comes out that where Creative don't release drivers for it.

Re:What I want from a motherboard...

[TheMeuge]

Agreed ... I still haven't found an onboard adapter better than my Live 5.1.

You mean like every single board that has audio based on the Via chipset that integrates the Envy24HT chip?

Live 5.1 is sonically one of the worst sound cards ever made. My 8-year-old Vortex2 from Aureal, is MUCH better... and the $20 Chaintech AV-710 absolutely blows it away.

Re:What I want from a motherboard...

[Aladrin]

It's simple: Adding onboard audio costs them almost nothing and gains them quite a few sales.

Re:What I want from a motherboard...

[TheGratefulNet]

look around for 'bit perfect' playback. you won't find it for soundblaster style cards. NOTHING from that company is even remotely pro audio quality or even home theater spdif bit-perfect output.

hint: their internal arch. resamples ALL data to 48k. even 48k gets resampled (man!, that's dumb) to 48k. hopes of having literal bit-perfect 44.1 is hopeless with creative brand.

envy24 - full-on pro chipset. I've used that one in my studio for years.

before that was the cmedia 8738 (still a gem if you can find it). also bit-perfect and has some great free drivers (sourceforge) that allow kernel streaming (win-xp) and bit perfect mixer-goof-proof output.

almost all else is drek. ie, junk.

Re:What I want from a motherboard...

[unfunk]

...and on that note, I have a SB Vibra 128 that I bought for $20aud about six years ago that has vastly, [i]vastly[/i] superior MIDI sounds than the vile DLS crap that my Audigy (and all other modern cards) uses. While I was studying Music Composition at uni, I actually ran both cards specifically so I could use the MIDI soundset of the Vibra.

Re:What I want from a motherboard...

[GreatBunzinni]

Touché. In fact, I've upgraded a while ago and I bought a nice little asus M2NPV-MX motherboard. I've chosen that particular motherboard for three simple reasons: it was a socket AM2 motherboard, it's price was very reasonable (about 60 euros) and it had an NVIDIA integrated video card (NVIDIA GeForce 6150). I don't play demanding games (mainly openttd), the integrated audio is excellent and from time to time I tinker with opengl. In the end that particular hardware combination made it possible for me to buy a new computer with a dual core processor for less than 200 euros.

So exactly what's wrong with integrated audio/video? Absolutely nothing. At least that's what my wallet's opinion.

Re:Call me ignorant, but...

[Svartalf]

You're not really that ignorant... :-)

In the case of the discrete cards (PCI-E, AGP...) they have a pool of memory that's accessible via the bus and that's directly accessible by the GPU's own memory bus (That memory size when you see 128, 256, 512Mb, etc.)- which is faster than just about anything out there and has no contention spots for the GPU to have to wait any longer than the access latency to the memory from the second access port. The peak speed of the GPUs when compared to an IGP solution comes from the contention-less, very, very fast access to the card's memory pool so that you don't stall the graphics pipeline. A stall of a microsecond can cost you FPS (Duh...) and larger stalls can drag framerates to the slide show domain- it's part of why the older ATI fglrx drivers were roughly 50% slower under Linux when compared to Windows. They had a stall in there somewhere that was introduced by their way of getting their then Windows-ish codebase to work under Linux.

Now, having said this, Hypertransport's suspiciously close to the same performance level of most of the local GPU buses and you only need to deal with bus contention issues for the only real performance snag. IGPs start making sense at that point for many applications because the memory's now close to the same speed as the add-in card's memory with similar latencies. The only real slowdown would be that you don't have dual pathways now.

Re:Looks like a fake

[PitaBred]

Do you actually read Slashdot? 90% of engineering types are apparently functionally illiterate, judging from the grammar and spelling atrocities that abound.