- Oct 9, 1999
- 72,636
- 47
- 91
Leadtek K7NCR18D PRO (nForce2) Review
- Thread starter NFS4
- Start date
Adul
Elite Member
Adul
Elite Member
I am not impressed with a couple of things...
1) he didn't test anything other then games and I usually discount most sandra, and pcmark benches anyways....
2) I am not impressed that we have to fully fill the ddr modules to get the best performance...I guess if I had 3 256mb modules but at this point that is unrealistic and the need to have 1 to 1.5gb of ram is also with my current 512mb module...
3) I am disappointed by the 400ddr scores....what is with amd mobos and running mem out of synch with fsb clock...I mean at least it should be no better no worse then the 333mhz ddr but it is a lot worse....I know that the cpu beocmes the bottleneck at 333mhz fsb with the pc3200 memory but it should just go unused..
sidenote: why does the ati9700 card not spank the ti4600 like it does in most benchmarks??? Is this a sign of nforce sabotaging competitiors cards on their boards and us ATI owners should beware of nvidia chipsets with ati vid cards???
1) he didn't test anything other then games and I usually discount most sandra, and pcmark benches anyways....
2) I am not impressed that we have to fully fill the ddr modules to get the best performance...I guess if I had 3 256mb modules but at this point that is unrealistic and the need to have 1 to 1.5gb of ram is also with my current 512mb module...
3) I am disappointed by the 400ddr scores....what is with amd mobos and running mem out of synch with fsb clock...I mean at least it should be no better no worse then the 333mhz ddr but it is a lot worse....I know that the cpu beocmes the bottleneck at 333mhz fsb with the pc3200 memory but it should just go unused..
sidenote: why does the ati9700 card not spank the ti4600 like it does in most benchmarks??? Is this a sign of nforce sabotaging competitiors cards on their boards and us ATI owners should beware of nvidia chipsets with ati vid cards???
sidenote: why does the ati9700 card not spank the ti4600 like it does in most benchmarks??? Is this a sign of nforce sabotaging competitiors cards on their boards and us ATI owners should beware of nvidia chipsets with ati vid cards???
The only game benchmark I see there is UT2003, Quake and Serious Sam and he didn't even say if FSAA or any other image enhancements have been enabled, an area where the 9700 would surely spank the ti4600 by a larger margin.
The only game benchmark I see there is UT2003, Quake and Serious Sam and he didn't even say if FSAA or any other image enhancements have been enabled, an area where the 9700 would surely spank the ti4600 by a larger margin.
>SpecView crashed almost every time using a GF4 Ti4600. ...
>we found the 30.82 Detonators to be the culprit. We use this particular version as it?s the latest WHQL certified non-beta
>driver released by nVidia on all motherboards,...
>Using the latest beta 40.72 drivers, we were able to solve this problem.
>we didn?t run into any problems with anything except the Creative MuVo MP3 player ... Although the device showed up
>properly along with directory listings, we couldn?t copy anything to or from this device.
I was informed in another thread that VIA chipsets suck because you sometimes have to change drivers to fix issues, so it looks like nForce2 sucks.
>this board should retail for approximately US$140
Kind of high for a mobo with a chipset that sucks. If they have an on-board video version for $80, it would be a good value (once the issues get fixed.)
>we found the 30.82 Detonators to be the culprit. We use this particular version as it?s the latest WHQL certified non-beta
>driver released by nVidia on all motherboards,...
>Using the latest beta 40.72 drivers, we were able to solve this problem.
>we didn?t run into any problems with anything except the Creative MuVo MP3 player ... Although the device showed up
>properly along with directory listings, we couldn?t copy anything to or from this device.
I was informed in another thread that VIA chipsets suck because you sometimes have to change drivers to fix issues, so it looks like nForce2 sucks.
>this board should retail for approximately US$140
Kind of high for a mobo with a chipset that sucks. If they have an on-board video version for $80, it would be a good value (once the issues get fixed.)
was informed in another thread that VIA chipsets suck because you sometimes have to change drivers to fix issues, so it looks like nForce2 sucks.
WTH is wrong with you? This happens to be a pre-production mobo. There are also better mobo makers around that would use the nforce2 like Asus, MSI and Abit so saying the nforce2 sucks because of one trial mobo made by a so-so mobo manufacturer is totally irrelevant.
Anyway, if you think nforce2 sucks, VIA still sucks the most
.
WTH is wrong with you? This happens to be a pre-production mobo. There are also better mobo makers around that would use the nforce2 like Asus, MSI and Abit so saying the nforce2 sucks because of one trial mobo made by a so-so mobo manufacturer is totally irrelevant.
Anyway, if you think nforce2 sucks, VIA still sucks the most
.This logic is sound
Via fans need to realize that until Via starts producing better chipsets on the FIRST try (kx133, kt133 -kt266, kt266a - kt400, kt400a) people WILL continue to consider them a second rate solution. Its sick to think on Nvidia's first two tries at mfg chipsets, they have soundly beaten VIA in performance, but more importantly Features:
Dual Channel DDR- Via hasnt even thought about it.
Respectable Onboard Video essentially free- Via doesnt feel like thinking about it.
Integrated NIC with minimal cpu usage- Via, only thinks about realtek
DD sound card integrated- Via lol, ac97??
Integrated branch prediction, hypertransport- Via, kt266 anyone....
Hmm, you're not trying to tell us something are you?
So when are we going to see a real nForce 2 review?
So when are we going to see a real nForce 2 review?
I'm not sure what you're trying to say, but many nForce implementations use a RealTek codec to access the APU.
>WTH is wrong with you?
WTH is wrong with me? This is a standard set by others, not a standard I adhere to. By the standard set to judge VIA, nForce2 sucks. No doubt the nForce2 will settle down in a few months (3 to 6) as they keep working on it. Typically for mobos, by the time they get it good, people talk it down on the forums because it is dated and they think there is another product over the horizon just about to kill it off.
This is a "near-release" mobo. Judging by the way info and ads were popping up in the last few days, nForce2 boards are immanent. I don't believe Leadtek should have any more trouble implementing the reference design than other mobo makers. Why do you suppose the nForce boards have a chip turned 45 degrees? Nvidia's reference circuit board traces have it that way. So the Leadtek board is probably like other nForce boards. As you may recall, according to the news, ALL nForce boards were postponed a while ago due to some problem in getting AMD's validation. That shows you how alike all the nForce boards are.
If you have seen the previous articles on nForce2, you recall that it is mostly a facelift for higher speed, and is nearly identical to nForce, and as such it is a design that has been around a long time, not really new. The big difference is the predictor for the memory controllers.
This reviewer did something unusual. He checked to see if the mobo worked with a few different things. L*rd knows why. But he did. What happened is just exactly what happens when you try different combos of hardware together: Some of them have "issues", some don't.
He ran "SpecView". If he hadn't he would never have known it would crash. This is exactly what happens when you run hardware, especially AGP video, with different combos of software. You get "issues." Nvidia has a beta out that fixes it. Probably SpecView is one of the issues Nvidia found out about.
What happens is the big video card makers tune their drivers to detect certain programs (like Quake3) so they will run faster. Years ago, the reviewers figured this out and began to find ways to get around it, so the manufacturers have to do it differently than they used to. The name Quake3 can't show up in the driver for instance. That's fine. But then some other program might have an unforseen issue because of the tuning or the detection. BTW, drivers are programs. So you can develop an interaction issue with a sound card driver, a USB driver, a bridge driver, or whatever. (The reviewer used a SB Live 5.1)
To give an idea how a driver MIGHT develop an issue: Suppose investigation shows that game One frequently does some things that never result in something different shown on the screen. A programmer develops a work-around so the video chip never has to process any of this, resulting in a faster game. The driver has a detection for this games behavior. Game Two does something similar but it would result in something different shown on the screen. So now Game two has an issue.
I have nothing against nForce. I built an nForce board, a cheapo ASUS AN266-VM. I was delighted at how nice this mobo was considering the price. I couldn't make it crash by stressing it, but then if I could have, it would have been a worthless POS. Inexplicably, no one but ASUS makes a mobo like this one (nForce 220), at least not one you can find. But I accidently ran into an issue. One keyboard that won't work with it. The keyboard is my ol' reliable, that has always worked before and still works with 4 other computers (I checked). Unlike every other component, I have never had an issue with a keyboard before.
WTH is wrong with me? This is a standard set by others, not a standard I adhere to. By the standard set to judge VIA, nForce2 sucks. No doubt the nForce2 will settle down in a few months (3 to 6) as they keep working on it. Typically for mobos, by the time they get it good, people talk it down on the forums because it is dated and they think there is another product over the horizon just about to kill it off.
This is a "near-release" mobo. Judging by the way info and ads were popping up in the last few days, nForce2 boards are immanent. I don't believe Leadtek should have any more trouble implementing the reference design than other mobo makers. Why do you suppose the nForce boards have a chip turned 45 degrees? Nvidia's reference circuit board traces have it that way. So the Leadtek board is probably like other nForce boards. As you may recall, according to the news, ALL nForce boards were postponed a while ago due to some problem in getting AMD's validation. That shows you how alike all the nForce boards are.
If you have seen the previous articles on nForce2, you recall that it is mostly a facelift for higher speed, and is nearly identical to nForce, and as such it is a design that has been around a long time, not really new. The big difference is the predictor for the memory controllers.
This reviewer did something unusual. He checked to see if the mobo worked with a few different things. L*rd knows why. But he did. What happened is just exactly what happens when you try different combos of hardware together: Some of them have "issues", some don't.
He ran "SpecView". If he hadn't he would never have known it would crash. This is exactly what happens when you run hardware, especially AGP video, with different combos of software. You get "issues." Nvidia has a beta out that fixes it. Probably SpecView is one of the issues Nvidia found out about.
What happens is the big video card makers tune their drivers to detect certain programs (like Quake3) so they will run faster. Years ago, the reviewers figured this out and began to find ways to get around it, so the manufacturers have to do it differently than they used to. The name Quake3 can't show up in the driver for instance. That's fine. But then some other program might have an unforseen issue because of the tuning or the detection. BTW, drivers are programs. So you can develop an interaction issue with a sound card driver, a USB driver, a bridge driver, or whatever. (The reviewer used a SB Live 5.1)
To give an idea how a driver MIGHT develop an issue: Suppose investigation shows that game One frequently does some things that never result in something different shown on the screen. A programmer develops a work-around so the video chip never has to process any of this, resulting in a faster game. The driver has a detection for this games behavior. Game Two does something similar but it would result in something different shown on the screen. So now Game two has an issue.
I have nothing against nForce. I built an nForce board, a cheapo ASUS AN266-VM. I was delighted at how nice this mobo was considering the price. I couldn't make it crash by stressing it, but then if I could have, it would have been a worthless POS. Inexplicably, no one but ASUS makes a mobo like this one (nForce 220), at least not one you can find. But I accidently ran into an issue. One keyboard that won't work with it. The keyboard is my ol' reliable, that has always worked before and still works with 4 other computers (I checked). Unlike every other component, I have never had an issue with a keyboard before.
-put down the crack pipe- you are delusional........
This is a PRE-PRODUCTION SAMPLE BOARD, explain to me how you know the "isssues" this board had were not related to leadteks implementation? Or Beta Bios. Almost all boards have issues early in their development, why do you think companys relesase bios updates?
KF, why don't you give it a rest?
Let's face it, VIA has a less than stellar reputation for quality, while it seems like most people who bought nForce based mobos are happy with them.
People aren't mindlessly bashing VIA for no reason, VIA brought it upon themselves.
For the recond, the last two computers I've built for myself have used VIA chipsets, and they work ok, but not as good as I'd like them to.
One is based on VIA's KT133, another on KT266A, both work fine for home usage, but they're not entirely stable, and considdering Im using high quality components, I blame this on VIA, since it seems like many other people are in the same "working fine but not as good as I'd like" boat.
nVidia by contrast seems to have gotten a pretty good reputation among DIY'ers.
Let's face it, VIA has a less than stellar reputation for quality, while it seems like most people who bought nForce based mobos are happy with them.
People aren't mindlessly bashing VIA for no reason, VIA brought it upon themselves.
For the recond, the last two computers I've built for myself have used VIA chipsets, and they work ok, but not as good as I'd like them to.
One is based on VIA's KT133, another on KT266A, both work fine for home usage, but they're not entirely stable, and considdering Im using high quality components, I blame this on VIA, since it seems like many other people are in the same "working fine but not as good as I'd like" boat.
nVidia by contrast seems to have gotten a pretty good reputation among DIY'ers.
- Oct 9, 1999
- 72,636
- 47
- 91
Originally posted by: Snoop
-put down the crack pipe- you are delusional........
This is a PRE-PRODUCTION SAMPLE BOARD, explain to me how you know the "isssues" this board had were not related to leadteks implementation? Or Beta Bios. Almost all boards have issues early in their development, why do you think companys relesase bios updates?
If I recall correctly, when the first nForce was released, it was riddled with bugs in the audio (which took a couple of driver releases to fix), had stability issues, and had the dual-channel DDR bug where performance dropped with a certain amount of DIMMs installed (I witnessed this firsthand with an Abit nForce board).
Yet people proclaimed how stable nForce was and how crappy VIA is.
DOUBLE STANDARD!
lol.. you crack me up.
WHY are you defending via? I guess you havn't suffered like the rest of us. Ever wonder why everyone is so excited about nVida producing a full featured chipset? It's nVidia's legacy to produce high quality products and support them with the best intentions. I'm not saying they are perfect, but I for one am sick of VIA and they need competition like this to make a better product.
WHY are you defending via? I guess you havn't suffered like the rest of us. Ever wonder why everyone is so excited about nVida producing a full featured chipset? It's nVidia's legacy to produce high quality products and support them with the best intentions. I'm not saying they are perfect, but I for one am sick of VIA and they need competition like this to make a better product.
If I recall correctly, when the first nForce was released, it was riddled with bugs in the audio (which took a couple of driver releases to fix),
The key word there is A COUPLE OF DRIVER releases. How many VIA 4-in-1's have been released? And it only took a BIOS update to fix nforce problems, VIA had to do HARDWARE revisions in order to fix problems(KT266 to KT266A, the famous 686B southbridge, now the crappy KT400 performance being replaced by the KT400A, etc..) . But despite BIOS and hardware updates, VIA chipsets still have lingering problems. Some problems are workable but generally, mobos with VIA will make you sweat more before you can 'tame the beast'.
But just to put things in perspective, just look at how long VIA has been in this game compared to nvidia and compare the number of issues and problems both have with their chipsets. Even compare VIA with SIS. There is no question that VIA based mobos have more problems. Compaq even felt comfortable enough to put nforce on a WORkSTATION model. If there is a chipset that would help AMD get the nod from PC makers such as Dell, it would be the nforce, not VIA.
The key word there is A COUPLE OF DRIVER releases. How many VIA 4-in-1's have been released? And it only took a BIOS update to fix nforce problems, VIA had to do HARDWARE revisions in order to fix problems(KT266 to KT266A, the famous 686B southbridge, now the crappy KT400 performance being replaced by the KT400A, etc..) . But despite BIOS and hardware updates, VIA chipsets still have lingering problems. Some problems are workable but generally, mobos with VIA will make you sweat more before you can 'tame the beast'
.But just to put things in perspective, just look at how long VIA has been in this game compared to nvidia and compare the number of issues and problems both have with their chipsets. Even compare VIA with SIS. There is no question that VIA based mobos have more problems. Compaq even felt comfortable enough to put nforce on a WORkSTATION model. If there is a chipset that would help AMD get the nod from PC makers such as Dell, it would be the nforce, not VIA.
This is a "near-release" mobo. Judging by the way info and ads were popping up in the last few days, nForce2 boards are immanent. I don't believe Leadtek should have any more trouble implementing the reference design than other mobo makers. Why do you suppose the nForce boards have a chip turned 45 degrees? Nvidia's reference circuit board traces have it that way. So the Leadtek board is probably like other nForce boards. As you may recall, according to the news, ALL nForce boards were postponed a while ago due to some problem in getting AMD's validation. That shows you how alike all the nForce boards are
Postponing a chipset to iron the kinks out is better than releasing a buggy chipset that gives headaches to the end user. And like you said, nforce boards are ALIKE but they are NOT THE SAME.
Postponing a chipset to iron the kinks out is better than releasing a buggy chipset that gives headaches to the end user. And like you said, nforce boards are ALIKE but they are NOT THE SAME.
It seems as though this would be the case. However, there is one factor you are not taking into account; that is latency. When things are running out of sync like that, it increases the latency. Latency has more to do with current applications than a high memory bandwidth. Therefore, it is hard to make a platform with higher latency performance better than memory with the same bandwidth and lower latency. Theoretically there should be more bandwidth, but the FSB limits that.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 21K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter