Is anyone else's Vnf4 ultra like this?

[Silversierra]

bump.

 

[Silversierra]

Originally posted by: Silversierra
If a 6600gt could reach 600mhz core, would it be similar to a 6800? Are there any benchmarks comparing 6600gt overclocked to 6800? What brand of low cost memory have people been using with the vnf4 ultra? I'm thinking of getting a gb, is mushkin ok? Corsair? Kingston Value? Are any dual channel 1gb kits single sided? My 512 kit is single sided, and it seems like people recommend single sided if you try to use all 4 dimms.

Any ideas?

 

[Chiefy]

Originally posted by: Silversierra
If a 6600gt could reach 600mhz core, would it be similar to a 6800? Are there any benchmarks comparing 6600gt overclocked to 6800? What brand of low cost memory have people been using with the vnf4 ultra? I'm thinking of getting a gb, is mushkin ok? Corsair? Kingston Value? Are any dual channel 1gb kits single sided? My 512 kit is single sided, and it seems like people recommend single sided if you try to use all 4 dimms.

I have the VNF4/Ultra with the blue PCB and new heatsink on the board. I also have a Jaton 6600 256MB Video Card. I also get the ' I tried using ntune, but it says to lock my pci bus' error as well.

To answer a couple of questions, no if you overclock the 6600 to 6600 GT clock speeds, the cards will not perform the same. The reason why is the 6600 GT has 1 more Vertex Pipelines and I believe the 6600GT uses DDR-3 Memory. So it turn out close, but not quite the same.

All the brands of memory you listed are good brands. I believe everyone's favorite though is Corsair. You might not get very good results from overclocking with the Kingston Value Ram. Maybe, maybe not, I'm sure many of the people here have some expericence with it.

I'm not new to PC's, but this board has puzzled me in a few ways. The way it handles RAM doesn't seem to make since and the whole "PCI Bus is handling HT" thing throws me way off. There have been some great posts in this thead and I appreciate how the anandtech forum's members don't make you feel like an idiot. These have got to be some of the best tech forums on the Internet.

 

[fr0thing]

I just received my VNF4/Ultra from Newegg. The date of mfg. is 12-29-2004 and the PCB is black but it has the newer fin-type passive heatsink on the NForce4 chip. I promptly popped off the heat sink, wiped off the factory thermal paste, and applied Arctic Silver 5. We'll see if that keeps the temps lower. I'm actually glad that I got the passive heat sink to keep the noise down.

Other than that the board itself was a breeze to install and started on the first try - which is pretty rare. It detected all of my hardware with no beeps or post errors, including a Raptor SATA. The board appears to be well made. I am not overclocking it until the next production revision of the BIOS is released. All of the speed settings are set to Auto. So far performance is very good:

SiSoft Mem Bandwidth test: 5625
3DMark03 (defaults): 9920

 

[tennesota]

Received my new board today. Mine is also the original dark brown pcb with the new heatsink [no fan]. I am replacing the stock heat sink with a Zalman northbridge heat sink.

 

[Gabb]

Wow! According to this, the boards with the passive heatsinks are not the newer revision- its actually vice-versa!

copied from somewhere!>>>>
Mainboard maker Chaintech said it had made certain improvements to its VNF4 Ultra VE mainboards following critics from X-bit labs web-site, who found certain disadvantages with the mainboard?s design.

Quote:
Mainboard maker Chaintech said it had made certain improvements to its VNF4 Ultra VE mainboards following critics from X-bit labs web-site, who found certain disadvantages with the mainboard?s design.

In the review entitled ?Chaintech VNF4 Ultra VE (NVIDIA nForce4 Ultra): Can You Have Good Hardware at a Low Price?? X-bit labs analyst Ilya Gavrichenkov wrote that the mainboard lacked active cooler on the media and communication processor, which potentially might cause instabilities.

?The chipset on this mainboard is covered with a passive heatsink of a medium size. Unfortunately, that?s insufficient as we will see later during our tests. The temperature of the chipset got as high as 60-70°C throughout the benchmarking process, which is hardly a normal operational mode for it,? Mr. Gavrichenkov wrote.

Additionally, X-bit labs reviewer criticized a number of BIOS issues the mainboard had, which ehnaces overclocking potential of the product series.

Chaintech informed X-bit labs on Wednesday that all the problems with Chaintech VNF4 Ultra VE have been addressed and the new revisions of the mainboard will be equipped with a fan and will employ a new BIOS version.

 

[MajorPayne]

Originally posted by: Gabb
Wow! According to this, the boards with the passive heatsinks are not the newer revision- its actually vice-versa!

copied from somewhere!>>>>
Mainboard maker Chaintech said it had made certain improvements to its VNF4 Ultra VE mainboards following critics from X-bit labs web-site, who found certain disadvantages with the mainboard?s design.

Quote:
Mainboard maker Chaintech said it had made certain improvements to its VNF4 Ultra VE mainboards following critics from X-bit labs web-site, who found certain disadvantages with the mainboard?s design.

In the review entitled ?Chaintech VNF4 Ultra VE (NVIDIA nForce4 Ultra): Can You Have Good Hardware at a Low Price?? X-bit labs analyst Ilya Gavrichenkov wrote that the mainboard lacked active cooler on the media and communication processor, which potentially might cause instabilities.

?The chipset on this mainboard is covered with a passive heatsink of a medium size. Unfortunately, that?s insufficient as we will see later during our tests. The temperature of the chipset got as high as 60-70°C throughout the benchmarking process, which is hardly a normal operational mode for it,? Mr. Gavrichenkov wrote.

Additionally, X-bit labs reviewer criticized a number of BIOS issues the mainboard had, which ehnaces overclocking potential of the product series.

Chaintech informed X-bit labs on Wednesday that all the problems with Chaintech VNF4 Ultra VE have been addressed and the new revisions of the mainboard will be equipped with a fan and will employ a new BIOS version.

Yeah, the original board (which I have) came with a passive heatsink that was fairly plain, and no markings, with a black/brown PCB. The second wave of boards have come out with either a blue or black PCB and a newer heatsink with the Zenith logo on it. This newer board/heatsink is not actually a new revision, just a slight update to the old board. The first real revision will come when chaintech starts sending out this board with active cooling on it, which is the one that x-bit labs was talking about. Some have asked if they will be able to exchange thier passively cooled boards for the newer rev board, and I have emailed chaintech about this, but have not (as yet) heard back from them.

 

[Silversierra]

What is the chipset listed as in the PC health section? Is it the "system" temp?

 

[ChineseDemocracyGNR]

bump

 

[Silversierra]

bump

 

[gobucks]

that would be my bet, silversierra, but i've been wondering the same thing. Mine is currently around 32-34C, that seems about right for a northbridge, right?

Also, has anyone had trouble running 1T command rate with this board? I can't get it to work with any memory, at any speed. It crashes Prime95 and memtest86, but at 2T I have the nice overclock listed below and it is stable overnight on Prime95. I can't figure out if it's a problem with the mobo or the CPU that's causing the problem, i hope it's just the bios, but we'll see i guess.

As for the newer bioses, they have fixed all kinds of problems for me - my memory used to always fail, but now it's stable, overclocking used to be a no-go, but now it's pretty damn decent (278 so far, won't hit 280, but that could be a CPU voltage problem). Also, the 1st 2 betas introduced 1T/2T timing control, but the option didn't actually do anything - it just defaulted to 2T. The newest (02/03) does fix this, but since I can't run stable at 1T, I can't vouch for whether or not it's working.

On a side note, does anyone know if installing the Zalman northbridge voids the warranty? I just bought one for 5 bucks, but I'm gonna wait to install it until I know I'm not gonna have to send my board back due to any problems.

 

[Silversierra]

My kingmax 2x256mb single sided cas 2.5-4-4-8 ram runs at 1T in the original bios.

 

[ShockAndAwe]

Hate to sound new, cause i've done quite a few forums of other kinds, but why do people type "bump"?

is it to get the thread back up onto the top of the que?

Thanx

 

[MajorPayne]

Originally posted by: ShockAndAwe
Hate to sound new, cause i've done quite a few forums of other kinds, but why do people type "bump"?

is it to get the thread back up onto the top of the que?

Thanx

Yeah, that is it exactly. They make a "bump" post to push the thread back higher in the forum. Some forums prevent this by either making "bump" posts a bannable offence (halflife2.net does this). Sometimes it can be a good thing tho, since some people are too lazy to look more than the top page deep into the forum, so any thread older than 1/2 an hour may never appear to them unless it gets bumped often. It is usually considered bad taste to do this to a very old thread too -- I have seen people go to the LAST page of the forum list, and just make a random reply to whatever the topic is just to put it back at the top and freak everyone out when they read some REAL old news, and think it is new!

 

[Ackbar]

Originally posted by: MajorPayne
Yeah, the original board (which I have) came with a passive heatsink that was fairly plain, and no markings, with a black/brown PCB. The second wave of boards have come out with either a blue or black PCB and a newer heatsink with the Zenith logo on it. This newer board/heatsink is not actually a new revision, just a slight update to the old board. The first real revision will come when chaintech starts sending out this board with active cooling on it, which is the one that x-bit labs was talking about. Some have asked if they will be able to exchange thier passively cooled boards for the newer rev board, and I have emailed chaintech about this, but have not (as yet) heard back from them.

Any news on getting a fan for us passive guys?

 

[Silversierra]

"Any news on getting a fan for us passive guys?"
Second that.

 

[rivethead]

Ok, it's been 10 days since the last post.....did anyone ever get one of those new boards with active cooling on it?

If so, how about a comment on the noise level.

I'm interested in this board, but not if it's going to be noisy. Personally, I'd prefer passive cooling, but not if there are problems!

 

[yokem55]

I have successfully installed an XP-120 on this board. I had to hook the one clip that interferes with the capacitor onto the new retention mechanism BEFORE screwing the second screw (nearest the rear connector panel) onto the board. Then with the clip hooked onto the retention mechanism, I rotated the rm into place and finished screwing it in and mounted the other clips on the RM. The mounting clip does push the capacitor in question aside sligtly, but not more than a couple of millimeters. Also, in contrast to what was discussed in the thread with the email from Thermalright, the first dimm slot IS available so long as your memory pcb is not too tall. The kingmax memory I purchased has a clearance of about 1-2mm between the heatpipes and the top of the memory pcb.

So far, I like this board. With the latest bios, it doesn't seem to be able to properly report clockspeeds with most settings. Occasionally it reports properly, but it does get wierd regularly. The cpu voltage adjuster is broken between 1.475 volts and 1.55 volts, anything below 1.6 automatically drops the voltage down to 1.45 volts. With 1.6v set, it works properly and runs at about 1.58 or so.

 

[GadgetBuilder]

There is a bit of humor in this cooling issue -- the Asus-SLI bunch are complaining about noise and installing passive chip coolers while the VNF4 bunch are complaining because they don't have active chip coolers.

Rivethead: You didn't say whether you intend to overclock. Overclocking is the real issue with the VNF4 board. The Xbit reviewer fried his board by noting the temp was above 60C and then raising the overclock -- not the best way to care for your hardware. He also didn't find the ability to control the fan speed which is available on the VNF4, just not in BIOS (you need to run SpeedFan).

http://www.xbitlabs.com/articles/mainboards/display/chaintech-nforce4_7.html

To be fair, read the whole review and judge the reviewer as well as the board.

I have the original VNF4 board, not OC'ed, and the temp runs 36C with the stock passive chip cooler. Running CnQ, the processor is 26C, 34C with load, the stock AMD fan runs 1740rpm at idle and is inaudible, with a barely audible whine at load when SpeedFan turns it up. Combined with a Sonata case, it is not silent but is quiet enough for me.

If you expect to overclock you will need more aggressive cooling than the stock passive cooler and should expect that noise will increase because of it.

Another thing to expect is some issues with NVIDIA drivers or SP2, hard to say which but they don't play well together on all NF4 systems, at least for now -- visit the NVIDIA mobo forums for insight on this.

 

[Ackbar]

Originally posted by: yokem55
So far, I like this board. With the latest bios, it doesn't seem to be able to properly report clockspeeds with most settings. Occasionally it reports properly, but it does get wierd regularly. The cpu voltage adjuster is broken between 1.475 volts and 1.55 volts, anything below 1.6 automatically drops the voltage down to 1.45 volts. With 1.6v set, it works properly and runs at about 1.58 or so.

Grr!!! I have the same problem!!! I hate this voltage problem, I can't believe that no one is really making a big deal about this. We really should be emailing Chaintech and demanding that they fix this, but everyone that I've talked to seems to just shrug it off. The volts between 1.45 and 1.6 (exclusive) are probably the most important ones and to not be able to use it is unexcusable in my opinion. I've gotten frustrated with this board lately... I basically only have 2 non-stock voltage options (1.45 and 1.6V) as I'd rather not go above 1.6 V. I hope they fix this issue soon or I'm going to jump ship and go with a different board.

 

[AnnoyedGrunt]

Hmmm, should have read this a few minutes ago.

Read the thread earlier, and the board seemed pretty good, so I ordered one about 1/2 hour ago. Should be here sometime next week.

One of the primary reasons I bought this board was because of the passive heatsink cooler. I'll be pissed if it comes with a fan. I think the Xbit guys were pretty lame to criticize the passive cooler as a "flaw", when for many of us it is a benefit of the board. Well, I'll have to see what mine is like when it arrives.

I guess if it is not the same as what is described, then I'll return it for a Soltek (which I also like quite a bit).

Are the voltage problems only for overclocking, or do they affect stock performance as well. I don't plan on overclocking, but if it affects stock performance or reliability then that's another story.

Thanks,
D'oh!

 

[Ackbar]

Originally posted by: AnnoyedGrunt
Hmmm, should have read this a few minutes ago.

Read the thread earlier, and the board seemed pretty good, so I ordered one about 1/2 hour ago. Should be here sometime next week.

One of the primary reasons I bought this board was because of the passive heatsink cooler. I'll be pissed if it comes with a fan. I think the Xbit guys were pretty lame to criticize the passive cooler as a "flaw", when for many of us it is a benefit of the board. Well, I'll have to see what mine is like when it arrives.

I guess if it is not the same as what is described, then I'll return it for a Soltek (which I also like quite a bit).

Are the voltage problems only for overclocking, or do they affect stock performance as well. I don't plan on overclocking, but if it affects stock performance or reliability then that's another story.

Thanks,
D'oh!

Nah, if you use 1.4 or even 1.45 V, you'll never notice any problems. The problem is that voltages between 1.45 and 1.6 (exclusive) V are selectable in the bios and using other programs such as clockgen, but they do not in actuality change the voltage. If you look in the bios or in CPU-Z or in clockgen or in nTune Monitor etc, you'll see that the voltages between 1.45 and 1.6V are physically unachievable. My board can not be set at any voltage between those. This is not a problem at all for people who:
A) are not overclock or
B) overclocking using higher voltages
Unfortunately, I fall in the middle so I'd rather use 1.5 V or 1.55V for example. If you notice most people using those board fall into A or B and hence never even noticed the flaw. I was hoping a new bios revision would fix the problem but that hasn't happened and now I'm starting to believe that it's a physical defect of the earlier boards. Hope that info helps!

 

[rivethead]

GadgetBuilder:

I haven't decided on whether or not I want to overclock. I do know I want a relatively quiet system. That's why I'm looking for passive cooling. Maybe Chaintech will put out two models: the Ultra OC (with active cooling for you OC people) and the Ultra with passive cooling. That would be nice, but I doubt their solution will be that simple!

I'm really attracted to the Chaintech due to its price/features, especially with the passive cooling.

My fear is that by the time I start buying and building (late March is my plan), the passive cooling units will be gone and nothing but the active cooling units will be left.

 

[tennesota]

Originally posted by: AnnoyedGrunt
One of the primary reasons I bought this board was because of the passive heatsink cooler. I'll be pissed if it comes with a fan.

If yours comes with a fan I'll trade you a passive heatsink cooler from another VNF4.

 

[GadgetBuilder]

Rivethead:

I spent 6 weeks collecting parts for my system, about half from eBay, the rest from Newegg and Outpost, so I know where you're coming from. I was concerned that if something went awry I'd be in trouble since the parts came from so many sources. Fortunately, I simply snapped the system together, turned it on and it booted up and ran Memtest86 off floppy (2 for $13 on eBay with no problems.

I used SATA drives for optical (Plextor 712SA) and hard drive (Seagate). I had heard rumblings of problems with SATA opticals on AMD systems so I had a "Plan B" ready to go with an IDE DVD. The Chaintech engineers apparently did their homework -- I put the XP installation disk into the 712SA, it booted and about 25 minutes later I was running XP. No fooling with moving SATA drivers to floppy, nothing special at all. I then installed NVIDIA drivers from the CD and proceeded from there. Most surprising to me was that XP could immediately talk on the net via cable with no effort or configuration on my part.

Overall, I started out wanting the ASUS NF4 ULTRA but when it started looking like vaporware and I was sitting there with a closet full of parts I shifted gears and bought the VNF4. Originally, I wanted the ASUS based on their record of stable boards; this VNF4 has been more stable than any other system I've owned so count me as a very happy camper on the hardware end of things.

The software is another story. The NVIDIA drivers and/or firewall and their interaction with XP are pretty shaky at present. This seems to be true across the board for NF4 systems. Not too surprising considering that NF4 boards are quite new. However, NVIDIA's recent 6.39 release doesn't seem to have been tested much with XP SP2 -- NVIDIA's reputation is that they typically take several months between releases so there could be a lot of grouchy NF4 users before the next release arrives - visit the NVIDIA forums to get a feel for what is going on.

Most troubling to me is there is no acknowledgment or recognition of problems by NVIDIA -- they seem to be insulated from the concerns of ordinary users. Perhaps the underlying problem is in XP and the driver or firewall is making the problem visible, either way it would be nice to know that the problem is recognized and a solution is being worked on.