Crashman

Polypheme
Former Staff
I don't see why THG ripped the nVidia, it did finish in the upper half of the boards, and all the rest had a the KT266A in common. The nForce boards even managed to win a couple benchmarks.
Those particular boards were aimed at the budget market, and are only the first examples of this technology. nVidia will undoubtedly release enhanced version of the drivers, and companies will release more performance oriented boards. A gain of only 5% would put the nForce above all the competition. All in all I think it's a great performing board, and would love to see more on the audio system.
Even if BIOS and driver improvements only manage to tie these boards with the best KT266A boards on the market, they will still be tied at the top. Good work nVidia, now what's your next revision going to bring?

What's the frequency, Kenneth?
 
G

Guest

Guest
I second that, Crashman. The benchmarks in Tom's article showed a board that was a little slower that the very fastest VIA boards, and very competitive with the rest of them. Considering that this is a first generation Athlon chipset from NVIDIA, and it's competing with a 4th or 5th (?) generation Athlon chipset from VIA... I say well done NVIDIA.

I assume that both Frank Volkel and Bert Topelt (the authors of the article) speak/write English as a second language, and I wonder sometimes if that makes some of their expressions come off a little different. In this and several past articles I have seen them use expressions like "can't hold a candle to", "trounced" and "pummeled" when speaking of differences of 2% or less. Not that I hold that against them--they know English a whole lot better than I know German. :smile:

One thing I did find a little disturbing about the article was the number of technical glitches, like quoting results different from what the benchmarks showed, or incorrect labels in tables. I hope these are fixed shortly so we can get the full picture.

Cheers,
Warden
 

Crashman

Polypheme
Former Staff
Don't kid yourself about these author's off the cuff comments. I know German and I know that they were getting out of hand on purpose. Either they have a love of VIA like so many of our forum members, or they have a beef with nVidia, or they thing this is what we want. In any of these situations they make themselves as arses, either by being biased, something not taken lightly in Journalism, or by underestimating their audience, an insult to our intelligence. Funny thing, I wonder how many members here didn’t notice?

What's the frequency, Kenneth?
 

WaferJ

Distinguished
Nov 27, 2001
5
0
18,510
I've notice that several articles lately have been like that - they seem to have their mind made up regardless of what the benchmarks show. In <b> Attack: Duron 1200 Takes On the Pentium 4 </b> for example I get the impression that they think the Duron 1200 is the best thing since sliced bread - It did beat the P4 in a few benchmarks, but it also finished dead last in several.. Don't get me wrong, I'm not an Intel fan and I wouldn't mind a Duron 1200 (have Celeron 300a right now :eek: till I uprgrade to XP1900+), but their statements are just too strong for their conclusions when the benchmarks don't quite follow.
And with the recent motherboard article they say "By and large, the VIA KT266A chipset knocks the stuffing out of the Nvidia nForce 420D." I for one don't think that 1 fps difference is knocking the stuffing out of it. I think if a board is 1 fps slower (out of 56 fps or 82 fps - just 2 examples) than another board, but has the features I want, I'll be happy with giving up 1 fps for those features...
Even before this article I was considering the nForce, and I'm even more likely to get one now... Oh yeah, and my company makes chips for nVidia = job security :smile:


"When all the chips are down, the buffalo is empty."
 

Crashman

Polypheme
Former Staff
LOL, the only reason I'm not buying one is because they haven't been around in Chipsets long enough to proove themselves. I'm leaning towards an XP333 unless something better comes along in a 735 board (not looking too likely).

What's the frequency, Kenneth?
 

louis

Distinguished
Mar 26, 2001
63
0
18,630
I read in the test setup that they used 2x128MB in the nforce boards. It would be nice to see some results with 2x 256MB of Ram on the nforce boards and 512MB in via boards, maybe the results could be a little different. Afterall, prices af ram aren't that high...
Good job from Nvidia, i'm going to get one for sure.
 
G

Guest

Guest
A few things to consider:

(1) Have they been running the benchmarks at the identical settings for all boards? If not, then the board that allows more aggressive settings will probably win. But there is no guarantee that it would be stable on those settings for 24/7. Aceshardware benchmarks all boards on identical, even if more conservative, settings.

For example, the MSI nForce board doesn't allow much memory tweaking (only CAS setting). Thus, if it were to be compared with other boards with all settings tweaked (and not just CAS), then it would be at a disadvantage. To be fair, bring ALL the boards back to the same settings.

(2) Have they been using the latest drivers? Perhaps, they might have been using pre-production beta drivers for nForce?

(3) Adding to my questions on the "pre-production phenomenon," it appears there is a contradiction. You can read on ASUS website that A7N266 allows raising FSB all the way up to 200MHz, not 158MHz as claimed in the article. Pre-production stuff again?

Some of the KT266A boards, too, seem to suffer. The article incorrectly states that the Shuttle board has FSB pegged at 133MHz, while this is obviously not the case.

Leo
 
G

Guest

Guest
I, for one, will probably wait a while longer before jumping on one of these new nForce boards. I've been waiting patiently since June to finally see what they could do, and I must admit I expected much better results. With a dual memory channel architecture, this board should have left the single channel KT266A far behind. Much like the i840 put the i820 board to shame. These new nForce boards cost quite a bit more than most KT266A boards out there. I already have an excellent sound card and a decent video card, so those extra features on the nForce boards would be mostly a waste.

But that's not the part that scares me the most though... the article mentions that the performance deficit might be fixed with proper drivers. But what if that's not the case? What if the problem is not in the drivers but in the chip architecture itself?

Let's think back to the original VIA KT266 chipset... it sucked the big one. Performance was poor. But people still bought it. How did those people feel when VIA fixed the problem in their KT266A chipset? None of those original KT266 board owners could upgrade their system in any way. There were no updated "drivers" to unlock this performance. There was no way to simply replace the chipset. The only choices were to either stick with their crappy board or buy a new motherboard.

I do not want to jump head first into buying a new nForce board only to find out a month from now that nVidia is releasing a "better" version of their chipset that blows away the first one. If fixing the drivers does enhance performance, then I'll be more than happy to dish out my hard-earned money. But for now, I think I'll just hang onto it.

--
Frank Gore
 

dhlucke

Polypheme
I'm not impressed. These boards were supposed to be better than sex. I just don't see what people have been waiting for. Maybe someone will release a better board, but I'm not going to hold my breath for too long. ASUS and MSI aren't exactly shoddy. Hopefully the next batch will be better though. It's nice to have nvidia in the market so that we get more innovation and competion.

I'm just happy I didn't wait any longer and went ahead and bought a SIS 735 board...

<font color=red>God</font color=red> <font color=blue>Bless</font color=blue> <font color=red>America!</font color=red>
 

phsstpok

Splendid
Dec 31, 2007
5,600
1
25,780
I still think nForce is great idea. Granted, performance isn't better than the best competing boards (at least not yet) and that is a let down. Built-in video, well it isn't worth mentioning and I'm glad THG didn't bother benchmarking. Like Crashman, I want see more about about the built-in sound. I also want to see more stability testing done but time will be the real test in that area.

The nForce boards seem tailor-made for OEM systems. All you have to put in them is memory and you have a business-class system. Put in the top processor and and you have system for a power user. Add to that a Geforce 3 card and you have a top-notch game system. Very attractive for system manufacturers.

I have to remind you we are still talking about excellent performance not like the old days of the horrible Intel 810 boards or the even worse integrated boards for Athlon/Durons. These boards not only had horrendous video performance (by design) but general system performance sucked, as well.

nForce could set a new bar.

<b>We are all beta testers!</b><P ID="edit"><FONT SIZE=-1><EM>Edited by phsstpok on 11/27/01 11:39 PM.</EM></FONT></P>
 

the_stranger

Distinguished
May 3, 2001
22
0
18,510
I just put together a system based on the MSI K7N240 nForce board for a customer who wanted a system for about $1200 CDN after tax, I don't think it would have turned out near as nicely with any other board and a separate video card and sound card, or some other crummy onboard like the horrific C-Media chip on my ASUS A7M266 (of course I'm using a Creative SB XGamer instead of that piece of crap).

Anyway, I don't have to deliver the system until next Monday, so I've been messing around on it a bit, it runs very nicely, and stable thus far (since yesterday). I did some Monkey's Audio encoding on it earliar today to heat things up for a half-hour or so and it remained stable. It zips around Windows XP quite nicely.

Anyway, I'll see how it goes in the coming few days... I'm presuming all will be fine. I'll post back if anything catastrophic occours!

--jeff
 

Ncogneto

Distinguished
Dec 31, 2007
2,355
53
19,870
I'm leaning towards an XP333 unless something better comes along in a 735 board (not looking too likely).
The leadtek board is looking like a dam nice board. All the right dividers to run your fsb at 166 and everything in spec, wonder how such an arrangement would fare against these two.

Another Cookie? Who is going to pay my dentist bill?
 
G

Guest

Guest
I want to see some change in the way benchmarking is done when testing the nForce boards. The dual memory controllers will come into play when the memory bus is put under stress by multiple simultaneous tasks. This is a serious difference in performance to the standard linear speed test. Let's see the VIA chipset hold up with multiple tasks, such as copying from one HDD to another AND over the network AND opening mail AND playing an mp3. Lets test this with IDE HDDs AND SCSI HDDs. This is far more relevant to us, now more than ever with XP handling multiple tasks at once WAY BETTER THAN 98 or ME. The linear benchmarks show very little in the real performance of the system.

The other thing is stability. VIA based motherboards are crap. Truely. The stability and compatability issues for VIA chipsets are SHOCKING and should never have been tolerated as they have been. Go and check out www.icronticforums.com if you don't believe me. See the enormous volume of issues posted there for VIA mobos. So, when reviewers start talking about real stability and compatability we can take them seriously.

Get your bloody testing right people!
 

tlaughrey

Distinguished
May 9, 2001
581
0
18,980
Some of the KT266A boards, too, seem to suffer. The article incorrectly states that the Shuttle board has FSB pegged at 133MHz, while this is obviously not the case.
I'm glad someone else caught the FSB mistake regarding the Shuttle board. I got a kick out of reading that as my Shuttle is running along at 137 MHz right now.

<i>Real knowledge is to know the extent of one's ignorance.</i>
 

tlaughrey

Distinguished
May 9, 2001
581
0
18,980
I agree that THG exaggerates a little too much. I read the caption for the article and expected to see the nVidia boards getting blown away. Then I looked at the benchmarks and I was like, what's the deal? This isn't the first article where I've noticed a discrepancy between the sensationalistic title and the actual benchmarks. Considering this is nVidia's first effort, I'd say they did pretty well.

<i>Real knowledge is to know the extent of one's ignorance.</i>
 

Ncogneto

Distinguished
Dec 31, 2007
2,355
53
19,870
Not exactly....newegg has had them for about a week now.

<A HREF="http://www.newegg.com/" target="_new">http://www.newegg.com/</A>

Another Cookie? Who is going to pay my dentist bill?
 

the_stranger

Distinguished
May 3, 2001
22
0
18,510
I'm with you on that one, which is why I keep buying Asus motherboards... they may not be the best performing but they have never failed me.

Unfortunately, I only have until Monday to mess around on this nForce system. Then the people who bought the system get to be the guniea pigs :)

*grin*

--jeff
 

DarkPhire

Distinguished
Mar 6, 2001
85
0
18,630
I'm getting an nForce this weekend :) , and one for xmas woo right on ;) lol

THG doesnt like Nvidia for some reason, they are totally skeptical about it. they still think nVidia is the devil for spoiling the beloved radeon 8500 release with the detonator XP's. LOL I love the radeon, but they need to get over it.

THG has released like 4 articles already on how much the geforce ti 500 sux, and the radeon 8500 rules. But somehow i dont understand why they take that spite with them to the Nforce. Oh well
I agree that THG shouldnt have been so hard on nvidia, but i think nVIDIA is doomed before they start sending cards to THG now.

--DarkPhire(United Devices team : AMDZONE.com; Member name : The Collective)
 

Ncogneto

Distinguished
Dec 31, 2007
2,355
53
19,870
<A HREF="http://www.vr-zone.com/reviews/Leadtek/7350KDA-OC/" target="_new">http://www.vr-zone.com/reviews/Leadtek/7350KDA-OC/</A>

Another Cookie? Who is going to pay my dentist bill?
 

pvsurfer

Distinguished
Jan 4, 2001
395
0
18,780
I still don't get the way nForce's Twinbank architecture utilizes those 3 DIMM slots. I've read that slot 1 is for the first memory bank while slot 3 is for the 2nd bank, so if that's the case, how does it utilize a DIMM in slot 2?

<b>God bless the <font color=red>U</font color=red><font color=white>S</font color=white><font color=blue>A</font color=blue></b>