nVidia nForce
cygnus writes: "Hardcore tech upstart site ExtremeTech.com got to check out an nVidia reference board, their findings are here. While most of the drivers and hardware were beta, they gave it some positive play. The story has a few large photos that are worth a gander... The reference board has only two PCI slots and no onboard Ethernet. Ouch. I'll stick with my Mac for now."
no onboard ethernet? perhaps you weren't reading very carefully... Connectivity: The MCP also has all the typical I/O connectivity and functionality you'd expect from a south bridge. These include two ATA/100 channels for hard drives, a hardware Ethernet 10/100 MAC and soft-modem support. The networking capability also supports HomePNA 2.0 (phone line networking). There's also the usual low-speed connections, including USB 1.1 support (two channels, for four ports), floppy disk support, PS/2 keyboard/mouse, and the legacy serial and parallel connections. The one thing missing from the picture is Firewire. nVidia chose not to implement 1394 into the MCP, which is an unfortunate choice. Windows Me already has robust support for 1394, and Windows XP will have a stronger implemention of 1394. In addition to connectivity with consumer devices, 1394 can also be used for small local area networks.
The REAL sam_at_caveman_dot_org is user ID 13833.
Option "NvAGP"
line in XF86Config. My card crashed after a few seconds of openGL, until I set this to "2". Now it still crashes, but it takes more like half an hour than a few seconds.
What they then go on to insinuate, though, is that nVidia is expanding into yet another realm of the PC hardware industry, which is, frankly, a scary thought to me. Not content with having a monopoly on the video card market, they are now going after the logic set market, intending to dethrone Intel in the biggest market of them all: low-cost motherboard chipsets.
What I'm really worried about here is the PC architecture turning into another closed, proprietary, overpriced piece of junk that the Apple has always been. When they start making video cards that are only "compatible" with nVidia chipsets, you will find out what I'm talking about. This is just the beginning of the market consolidation for nVidia. First 3DFX and now this. You can already tell that they are against open standards when they refuse to release open source drivers for their newest cards so that they can be supported under Linux and other free operating systems. What do they have to hide, anyway?
While no one on Slashdot may be interested in buying a motherboard manufactured with this chipset, there will be a lot of people who will. While one may even get decent performance out of the chipset, perhaps even better than the Intel eqivalent, one must consider: is it really the best choice that I can make in the long run? My personal opinion is that the answer is no, it ain't worth it.
--
"I don't know that atheists should be considered citizens, nor should they be considered patriots." - George Bush
I know this one is still in beta so I can't expect too much, but can anyone tell me why so many of these video card reviews never have screen shots? I don't care if the card goes fast if it doesn't look good.
Sigi
"You can justify anything by putting it in quotes, adding a famous name and making it a sig" - Albert Einstein
So I assume when you say it only has support for 2 PCI slots you mean, "It supports 5 PCI slots, just like all the other AMD-supporting chipsets, but this one board only implemented two of them, which is irrelevant to us because it's merely a reference board and we'll all buy the Asus full ATX that they've already designed," right?
This is a self-referential sig
One neat thing that this chipset supports is the so-called "twin-bank" technology. This interleaves access to the DDR DIMMS (ie. byte 1 is on DIMM1, byte #2 in on DIMM#2, byte #3 in on DIMM #3) to pump up the maximum memory bandwidth beyond what a single DDR DIMM would be capable of. I guess they had to do this since the GPU uses system memory (ugh), but it has a nice side-effect of really unleashing the performance potential of 1ghz+ CPU's, especially if you ditched the onboard video and stuck something decent in the AGP slot.
;)
And to think, my 386 had interleaved memory back in the day..I was 'leet and didn't even know it
AMD recently claimed they would have 30% overall market share by the end of the year, their goal before launching their 64 bit processor. It seems they are on track to meet those expectations. With NVidia's chipset offering a low-cost affordable solution for low-end machines and the 760MP chipset offering us the world's first multiprocessor AMD platform, they are putting themselves in a position to have real sway in the coming 64 bit desktop revolution.
120 characters isn't enough to explain it.
This puts it in a strange position in the market. The chipset is very powerful, yet the graphics will be decidedly average when the chipset is finally being sold on the market. It beats all the other integrated systems out there (the audio system is to die for, beating the SBLive! into a cocked hat) by a massive margin, but they only cost like $20 a pop anyway!
The SiS735 looks to be the other chipset to look out for, beating the VIA and AMD solutions at the moment, and being a much cheaper single chip solution.
Both chipsets incorporate next generation interconnects - nForce uses an 800MB/s Hyperlink connection, and the SiS uses a 1.2GB/s multi-threaded connection (possible due to the single chip design).
VIA will soon be releasing their second shot at a DDR chipset for the Athlon, called the KT266Pro, without the problems the first one had, and improved DDR memory interface.
Whatever, the future is looking great for the Athlon in terms of chipset support, which now covers the entire low to medium end of the market (up to workstations and low-medium end servers with the 760MP).
If they intend to take on Intel's i810 and i815 successes, then why do it with a chipset for an AMD CPU instead of one for Intel's? Sure AMD is getting bigger and bigger, but I haven't seen any marketshare survey give them more than 30% at best. And with the current largest PC maker in the world (Dell) not selling Athlon's at all they're severely limiting the reach of this new chipset it seems.
Seriously, it's great that motherboard-makers like Asus are going to integrate the nForce, but what's the use? People building their own PC (the ones buying Asus/Abit/MSI/etc. boards) probably don't want to get an integrated GeForce2 MX...
So anybody know what the deal is? Licensing?
Now nVidia is going to make it possible to make an integrated motherboard, and the performance is going to be excellent. The Duron totally crushes the Celeron, the GeForce MX totally crushes the 810 onboard video, and the audio DSPs totally crush everything currently on the market. As long as the price for a Duron plus one of these boards is about the same as the price for a Celeron plus an 810 motherboard, they will sell a whole bunch of these.
steveha
lf(1): it's like ls(1) but sorts filenames by extension, tersely
You could at least say something in the writeup about what the nForce is - is it a new graphics card? A motherboard? A chipset? A toaster oven? I know almost everyone else follows hardware development religiously, but I had no clue what this post was about. The linked article didn't help much at first since it assumed the reader already knew what was being reviewed.
The elipses in your quote, "... no onboard ethernet", refer to, in part, the reference board. The article does not state, contrary to what you way, that the reference board has on-board ethernet. The fact that the chipset supports it is another matter. But then you knew this, given the last paragraph in your reply in which you draw the distinction between the number of PCI slots on the ref board and the number supported by the chipset.
Too bad you haven't bothered to read about the nForce before commenting. The nForce supports an AGP slot, and in fact in the article they mention putting a better video card in with no problems.
Just because the chipset has built in ethernet, it doesn't mean that there is an ethernet connector on every board with that chipset. Looks like this reference board doesn't have the digital audio connector(s) either. In short this board is missing quite a few of the features that are supported by the chipset, but I'm sure asus or someone will make a more full featured version for their board.
This sounds great to me. Small M/B, reasonable graphics, sound etc. Mix with a wee power supply and a laptop HDD and I'll have a PC that I can comfortably carry to work & back. Personally couldn't care less if it had NO PCI slots, but it must have ethernet.
A couple of days ago there was an article here on /. discussing the nature of content the ExtremeTech.com website. Some comments on that article at the time (including mine) made the mistake of assuming that the site was being run by ZD Net, which is a subsidiary of C|Net, which most readers here look down upon, however, aparently that information was in error. The tite is in fact managed by ZD Media which is a new division of Ziff Davis Publishing.
/. article is defintely worth a read though, when evaluating content from the Extreme Tech website.
The earlier
Having said that, and regardless of the earlier postings here indicating inacuracies in the ExtremeTech hardware review being discussed here, I have to say the site look quite well put together and editorially fair.
--CTH
---
--Got Lists? | Top 95 Star Wars Line
...with a smattering of "I'm a real hard core tech site" pasted on front. (Modding myself down 1 point.)
A firewall can not protect you from yourself. Turn off what you do not need. Do not use the firewall to do your work.
The reference board has only two PCI slots and no onboard Ethernet. Ouch. I'll stick with my Mac for now.
Do you think this stuff could really compete with mac anyway?
Compete is a relative term, depends on the task (say, Photoshop vs Quake). I'm interested to learn how the onboard gfx of nForce compares to say, a GF3 in a Mac or PC. GF3 may be a bit faster, but is limited by AGP 4x/6x/8x and the system ram thruput. I'd also like to see if NVIDIA is working on some sort of unified color correction API for Windows (a la Apple's ColorSync... something that is universal, not just one part of a kluge of components).
The crossbar architecture reminds me a lot of my Silicon Graphics Octane... built in January 1997. As far as wide memory bitpaths, I think the Sun Blade 1000 is the highest for less than $12,000 --288-bit ram path (upgrade four dimms at a time).
At any rate, I think I'll wait for the SiS735.
I use a bit of everything and while I only own one mac these days, we use almost nothing but G4s at work (FinalCutPro 2 with DVcam and DVCPro gear). At home I spend most of my time on a few old SGI workstations but slowly using more and more Mac OS X on my PBG3. As soon as Apple finishes (or starts, heh) full hardware acceleration for the Aqua widgets, I'll probably go fullbore OS X and never look back. There's something very nice about using Photoshop, postgresql, apache, and iMovie all on the same machine under the same os. I do admit that OS X hasn't been perfect, but our Mac OS 9.1 machines at work have been rock solid.
This puts it in a strange position in the market. The chipset is very powerful, yet the graphics will be decidedly average when the chipset is finally being sold on the market.
Why is that? I mean, if it's very powerful, the graphics should be more than average, right?
It beats all the other integrated systems out there (the audio system is to die for, beating the SBLive! into a cocked hat) by a massive margin, but they only cost like $20 a pop anyway!
Well, better technology should be more expensive, don't you think?
Being a Mac user (as well as a NeXTSTEP/Openstep, Solaris, IRIX, OpenBSD, Linux, and Windows user), I have indeed run across far too many "typical mac users". Half of which are addicted to the platform and refuse to even look at anything else... the other half not knowing the difference between USB and HBO (or IEEE1394 and the IRS 1040).
I use a number of G4s at work as they've been powerful, stable, wonderful machines to work with for DV video editing and compositing (under Mac OS 9.1). The G4 case was a deam to work with when adding ram and a second hard drive to each machine. We currently have two SGIs and three PCs running Maya but will certainly try Maya under OS X later this summer. At home and for some of the organizations I work with, I use a variety of x86 and workstation platforms. I do my gaming and game serving on two homebrew Windows PCs. No need to bash a platform or try to convert the dissimilar. Use what you enjoy and what works for you.
It's hard to tell exactly what the chipset can do, as opposed to what the reference board does. I would assume that, if this chipset can't handle a better GPU, they'll make one before long that can. Likewise, most of the obvious problems are probably a result of them doing the motherboard when they normally do chipsets.
If they've improved memory bandwidth enough that the GPU and CPU can both use the same memory without getting too slow, that would probably be a big advantage-- you could dynamically allocate memory size and memory bandwidth, so that you only give the graphic system a huge amount of memory when it wants it, and you can use the memory for programs when you're not doing that much graphics.
Why is that? I mean, if it's very powerful, the graphics should be more than average, right?
No, the graphics won't be more than average - the GeForce 2MX, which is what Nvidia has integrated onto this motherboard, has been their value graphics chip for some time now. It isn't nearly as fast as GeForce2, GeForce2 Pro or GeForce3, and the motherboards aren't even on the market yet.
What the post you replied to pointed out, is that this is a strange combination - while other features on the motherboard, like the built in sound and memory architecture are high end, the graphics aren't in the same class. So many (myself included) would like to have the same chipset, without the builtin graphics - no point in having that chip if you want to buy a top-of-the line graphics card anyway.
The chipset's North Bridge- Dubbed the "IGP", it provides for more efficient use of memory bandwidth using the GeForce3's crossbar switching technology. Note that this is the same kind of technology that SGI, Sun, and other high-performance UNIX hardware vendors have been using for their memory architecture (drool...)
Support for 128-bit DDR memory- This is just amazing. Not only will their particular implementation allow you to put different sizes of DIMMS in the board, but it will also allow DDR memory to boost it's bandwidth 100%, which means that instead of maxing out at 2.1GB/sec, it maxes out at 4.2GB/sec (again, drool...)
The link between the North and South Bridges- This will support bandwidth up to 800MB/sec, which means that it will have an excess even under unusual conditions, such as maxing out ATA/100, maxing out the PCI bus and outputting 256 streams of digital audio all at the same time. Simply awesome.
The built-in audio- The article gives a lengthy description of this. Suffice to say, it is better than any consumer card available today, built-in or otherwise.
Built-in graphics system- This actually was a little disappointing to me, because, while they may have it running at essentially AGP 6x, they used their budget system, the GeForce2 MX, for the processing. I'm also disappointed that it uses a "shared video memory" architecture that can be found in a lot of deplorable video platforms, notably the Intel i810 chipset. Basically, it uses up to 32MB of system memory for the frame buffer in addition to its onboard memory. However, with the huge amount of bandwidth available to the system memory, this may or may not be an issue.
Lack of support for 1394 and 64-bit PCI- Regrettably, they don't seem to have included support for either Firewire or 64-bit PCI. The lack of 1394 support is particularly surprising, as it has gotten increasingly better support from both Windows ME and Windows XP, as these are the main platforms which will be running on this board.
All in all, I think this is going to be an awesome board, a real leap ahead of everything else out there for the x86 market right now, at least in the consumer arena. I look forward to buying one, if the sticker shock isn't too harsh ;-).
--
< )
( \
X
Friends don't let friends use multiple inheritance.
Another interesting problem is the issue of memory bandwidth. There is some concern, rightfully, about available system memory bandwidth when using the integrated graphics core. In theory, the CPU should always be able to grab memory bandwidth first. When using the built-in IGP graphics, the memory streaming tests came in about 15-20% lower than when a GeForce2 Ultra was dropped into the AGP slot. This is possibly related to their memory arbitration logic still being enabled and consuming cycles even when nothing else is truly vying with the CPU for memory access.Another interesting problem is the issue of memory bandwidth. There is some concern, rightfully, about available system memory bandwidth when using the integrated graphics core. In theory, the CPU should always be able to grab memory bandwidth first. When using the built-in IGP graphics, the memory streaming tests came in about 15-20% lower than when a GeForce2 Ultra was dropped into the AGP slot. This is possibly related to their memory arbitration logic still being enabled and consuming cycles even when nothing else is truly vying with the CPU for memory access.
Exactly how does the reviewer think the picture is getting on the screen if the RAMDAC isn't accessing main memory? Even suggesting that the CPU should have higher priority than the RAMDAC is either ignorance or stupidity. Of course the RAMDAC has to have priority or you are going to end up with big blank patches on the screen!!
Fear: When you see B8 00 4C CD 21 and know what it means
Actually, for the longest time, everyone has been shipping integrated graphics with AGP slots. HP used to not do that, and they caught hell for it. I doubt you'll find many computers at Best Buy that have integrated graphics and no AGP slot.
A deep unwavering belief is a sure sign you're missing something...
Check your BIOS: If you have a setting called "Memory Interleave" or something like that, you likely can do this with your motherboard today. The motherboard I have that does this (an EPOX EP-MVP3G) has three settings: None, Two Bank, and Four Bank. Benchmarks suggest a major improvement with a K6-2/450 when Two Bank is selected, although I don't have enough slots to try Four Bank out (it just reverts to two, or so it seems).
And dareth I mention it, many Macintoshes have been able to do this for quite some time now. Just add extra RAM to a Mac, make sure all the SIMMs are identical, and you will suddenly have a nice performance boost for more than one reason.
Err, correct me if I'm wrong, but the 'onboard' video card still has ram on it. Probably something like 8 or 16mb or so (maybe 32 if they were nice). Enough for a few framebuffers. The main memory is used to store textures that the video card can't hold. This is the main idea behind AGP that is now conveniently forgotten because cheap and performance are two opposite goals...
OK, I'll correct you ;)
The onboard video DOES NOT have integrated frame buffer memory. This is the same for most i810's, i815's etc. In fact, you can lower system performance by running your monitor at a higher refresh rate! To find out how much memory bandwidth your monitor is using on these things, mulitply your screen number of pixels (width * height) by 2 for 16-bit color or 4 for 24/32 bit color. This will give you how many bytes per refresh the RAMDAC is using. Multiply that by your refresh rate (ex. 85) and you have bytes per second. Take this off of your maximum memory bandwidth, and that's what you have left for your CPU. Not a great situation!
Dan
_damnit_
_damnit_
It's my job to freeze you. -- Logan's Run
The frame buffer is part of main memory - you set up how much main memory to use as the frame buffer in your BIOS.
Sure it has an "AGP 6x" connection, but this doesn't help a lot when you are pumping all the data to the RAMDAC through that connection as well.
Fear: When you see B8 00 4C CD 21 and know what it means
Because nVidia is secretly in cahoots with AMD. They are currently developing a GeForce4/Athlon "system on a chip" which will be the worlds first processor to have world class graphics built in.
Thinkaboutit.
More