Soltek NV400-64: Single Channel nForce2 400 Review

[DAPUNISHER]

It's only 76$ shipped at newegg and is faster than the Ultra400

This is also interesting in the fact that it?s the first time we have seen anything based on the nForce2 400 chipset. Since we never received a reference board from anyone, we are taking the Soltek board as the defacto performance we can and should look for on future nF2 400 motherboards. Which raises the question: why is the nF2 400 performing better than the nF2 Ultra 400 board? Too be completely honest, I am not sure and any answers I have gotten from NVIDIA seem stretched at best.

Link

 

[First]

Originally posted by: DAPUNISHER
It's only 76$ shipped at newegg and is faster than the Ultra400

It can't be faster than nForce2 Ultra 400 if the NV400-64 is based on the nForce2 400 (single channel) chipset.

 

[EY2K]

Originally posted by: Evan Lieb

Originally posted by: DAPUNISHER
It's only 76$ shipped at newegg and is faster than the Ultra400

It can't be faster than nForce2 Ultra 400 if the NV400-64 is based on the nForce2 400 (single channel) chipset.

According to the benchmarks in the review, it IS faster. and so is the reviewer.

 

[DAPUNISHER]

Originally posted by: Evan Lieb

Originally posted by: DAPUNISHER
It's only 76$ shipped at newegg and is faster than the Ultra400

It can't be faster than nForce2 Ultra 400 if the NV400-64 is based on the nForce2 400 (single channel) chipset.

I thought as much too, but either he goofed up his methodology somehow or the numbers speak for themselves. Perhaps you could request one to review and see if it was a fluke/goof or not.

 

[Insane3D]

WTF?? That is weird...

Evan, can you inquire with Ryan over @ AMDMB and see if you can duplicate his results?

 

[First]

No disrespect to Ryan but I think he must of not isolated his testbed properly. NVIDIA would be stupid to knowingly sell faster than U400 chipsets at non-U 400 chipset prices and let the online media know about it. Once I get a hold of a non-U 400 board I'll let you know my results.

 

[mechBgon]

Even if it were 5% slower than nF2 Ultra 400, that could still be a nice well-rounded board. With the 8RDA+ below $90 and the non-Ultras coming, the AMD crowd has a lot to be happy for

By the way, I see EPoX has coughed up a whole bunch more nForce2 boards: http://www.epox.com/html/motherboard-products.asp?socket=socket%20a&lang=1

:Q ~ whoa!

 

[NesuD]

Well he doesn't actually say whether or not he ran the tests on the A7N8X in a dual channel memory configuration or not. I would expect that he would but he just doesn't say one way or the other.

 

[txxxx]

Maybe AMDMB should run a memory latency test? Could be the single channel system on the NV400 standard chipset has an overall memory access latency. Although from what I remember, DUAL channel created a lower latency.......mmmmmmmm

 

[yodayoda]

wow, that is an eye-opening review. i wonder if he had all the timings set right and if he had the Asus in dual channel configuration. if that is true, snatch up a NF2 400 before nVidia stops them from going out into the retail channel, as they will kill their ultra400 (and their ultra profit margin) board sales =)

 

[DAPUNISHER]

Originally posted by: NesuD
Well he doesn't actually say whether or not he ran the tests on the A7N8X in a dual channel memory configuration or not. I would expect that he would but he just doesn't say one way or the other.

If you look at the test system setup on pg3 of the review it shows 2x256mb Corsair XMS3500 is used for both boards. He also says the following about the setup for those asking about the timings and DC-DDR,

I tested the Soltek board with a single DIMM as well as with two DIMMs, and the scores were close enough for the difference to be negligible. Because, again, the NV400 is only a single channel memory motherboard, using one or two DIMMs makes no really performance difference. The same cannot be said of the nF2 400 Ultra in the Asus A7N8X board. All the memory timings were kept the same for each board (6-2-2 2.0)

then in the last SPECviewperf 7 he states

Finally, some success for the A7N8X motherboard. The tests that are in the SPECviewperf 7 suite are closely related to workstation class applications, and their hunger for memory bandwidth gives the upper hand to the nForce2 Ultra 400 chipset here, and by a pretty noticeable margin, with the exception of the final Unigraphics set.

So it looks like bandwidth intesive tasks will give the decisive advantage back to dual channel, and that coupling the IGP with the sc 400 would be pointless, but I'd love to see a review by another site, and Evan in particular to see if the results can be duplicated.

 

[Ketchup]

Originally posted by: Evan Lieb
No disrespect to Ryan but I think he must of not isolated his testbed properly. NVIDIA would be stupid to knowingly sell faster than U400 chipsets at non-U 400 chipset prices and let the online media know about it. Once I get a hold of a non-U 400 board I'll let you know my results.

If he indeed neglected to properly "isolate his testbed" he must have done so after he took the PC-Mark scores, because the difference in CPU and Hard Drive scores is negligible, leaving the higher memory score as a good indication of where the performance on this board comes from. Same story for Sisoft Sandra.

 

[NesuD]

Ok DP all indications are that the asus ran a dual channel configuration. I would say it just proves how very little difference Nforce2 dual channel makes in general. Looks like the dual channel only showed it's muscle in a test that benefits from memory bandwidth. Almost seems as if Soltek was able to optimize the single channel memory performance once freed of having to design for a stable dual channel memory controller.

 

[DAPUNISHER]

Originally posted by: NesuD
Ok DP all indications are that the asus ran a dual channel configuration. I would say it just proves how very little difference Nforce2 dual channel makes in general. Looks like the dual channel only showed it's muscle in a test that benefits from memory bandwidth. Almost seems as if Soltek was able to optimize the single channel memory performance once freed of having to design for a stable dual channel memory controller.

Yes sir, it certainly does begin to look that way but I'll withhold any further speculation on the matter until more reviews are done. I will say that if your positing is correct then the 400 chipset could be given a better feature set than appears on the Soltek to act as a less expensive solution to steal more thunder from the KT600.

 

[Insane3D]

When are we going to see the new Nvidia SB with native S-ATA, the MCP-S? My only gripe with Soltek is they seem to refuse to put out a higher end board with all the MCP-T goodies..

 

[DAPUNISHER]

Originally posted by: Insane3D
When are we going to see the new Nvidia SB with native S-ATA, the MCP-S? My only gripe with Soltek is they seem to refuse to put out a higher end board with all the MCP-T goodies..

Yeah, that's my only complaint with them too but I think their idea may be to break in with great overclocking and stability at a very nice pricepoint then when they have a better following target a lager market perhaps? Otherwise the lack of the MCP-T makes little sense

 

[MARMADUKE]

No disrespect to Ryan but I think he must of not isolated his testbed properly. NVIDIA would be stupid to knowingly sell faster than U400 chipsets at non-U 400 chipset prices and let the online media know about it. Once I get a hold of a non-U 400 board I'll let you know my results.

Ryan used the worse U400 mobo on the market!.. Asus A7N8X 2.0..but it has soundstorm!..WOW!
All my Asus A7N8X 2.0 customers cannot get this mobo up to par? the Asus A7N8X 2.0 is like a KT400 with soundstorm for only $149 :Q

Ryan isolated a deathbed for the Asus A7N8X 2.0 properly!
I'm sure the $75 Soltek NV400-64 gave the Asus A7N8X 2.0 with soundstorm..WOW!.. the beating it deserved.

I did almost the same exact tests with my trusty $77 Epox 8RDA+ 1.1 crush with soundstorm
200FSB DC
CPU-same
memory-2X256 TwinMOS CH-5.. timings 6322

The Soltek NV400-64 keeps very close but never passes the Epox 8RDA+ @ 200fsb!.. but once you go into Epox & Abit territory 210fsb-250fsb.. also known as non-Asus territory, the mighty ULTRA 400 DC power of the Epox 8RDA+ will bring the Soltek NV400-64 to it's knees!
This is why Nvidia said that U400 is KING!

 

[DAPUNISHER]

LMAO@MARMADUKE Why not tell us how you really feel? I have to point out though that the Soltek's overclocking ability wasn't explored and if it holds to the same criteria as the FRN2 is capable of then it would give the Epox and Abit a good run

 

[MARMADUKE]

DAPUNISHER
over 200FSB and the DC comes in to play big time!...the soltek would be overcome with DC bandwidth!

Yea..but a a few bucks more you can buy the KING!.. 8RDA+ 2.0 @ newegg with bios VDD and soundstorm

 

[KF]

>over 200FSB and the DC comes in to play big time!...the soltek would be overcome with DC bandwidth!

don't see why or how. With the CPU and memory locked to the same FSB, speed increases in direct proportion to the FSB, whether dual channel or not. It has to.

 

[billandopus]

Originally posted by: DAPUNISHER

Originally posted by: Insane3D
When are we going to see the new Nvidia SB with native S-ATA, the MCP-S? My only gripe with Soltek is they seem to refuse to put out a higher end board with all the MCP-T goodies..

Yeah, that's my only complaint with them too but I think their idea may be to break in with great overclocking and stability at a very nice pricepoint then when they have a better following target a lager market perhaps? Otherwise the lack of the MCP-T makes little sense

Lager market? Whoa .... wait up, Soltek is gunning for alcoholic computer enthusiasts market!

 

[DAPUNISHER]

Originally posted by: bill_n_opus

Originally posted by: DAPUNISHER

Originally posted by: Insane3D
When are we going to see the new Nvidia SB with native S-ATA, the MCP-S? My only gripe with Soltek is they seem to refuse to put out a higher end board with all the MCP-T goodies..

Yeah, that's my only complaint with them too but I think their idea may be to break in with great overclocking and stability at a very nice pricepoint then when they have a better following target a lager market perhaps? Otherwise the lack of the MCP-T makes little sense

Lager market? Whoa .... wait up, Soltek is gunning for alcoholic computer enthusiasts market!

LOL! I'm not only a member of the alcoholic computer enthusiasts association, I'm the president

 

[NesuD]

check this socket A chipset roundup. They compare an abit board with the nforce2 u400 with a the soltek single channel board as well as the new via kt600 chipset and the newest SIS socket A chipset. Their results clearly put the U400 on top of the single channel 400 although it is by a pretty slim margin in most tests. Nforce2 anything basically manhandles the rest hands down. Socket A chipset roundup

 

[DAPUNISHER]

Originally posted by: NesuD
check this socket A chipset roundup. They compare an abit board with the nforce2 u400 with a the soltek single channel board as well as the new via kt600 chipset and the newest SIS socket A chipset. Their results clearly put the U400 on top of the single channel 400 although it is by a pretty slim margin in most tests. Nforce2 anything basically manhandles the rest hands down. Socket A chipset roundup

I thought that at first glance too but if you look closer

For NVIDIA nForce2 Ultra 400 we selected ABIT NF7 2.0 mainboard. This solution demonstrates almost the fastest performance today among all mainboards based on the same NVIDIA core logic. For this reason and also because the mainboards based on the single-channel nForce2 400 are not that widely spread yet, we used the same mainboard in single-channel mode to test the single-channel NVIDIA chipset.

Until we see more reviews using the NF2 400 SC and not a ultra400 in SC mode it's still not possible to say conclusively that the ultra400 is faster even though all the known spec indicate that should be the case.

 

[MARMADUKE]

Lets do a REAL review!.. right here.. ! right now..!
my $77 8RDA+ with soundstorm @ 500Mhz DC mode (no VDD mod) VRS any $149 A7N8X 2.0 mobo with defective SATA controllers... LMAO