My new rig rocks :D

[ndee]

OK, my mobo died last sunday and so I bought a new mobo/cpu/ram combo. I got a Asus A7N8X Deluxe, Athlon XP 2400 and a stick of 512MB PC2700 Kingston Value RAM. I installed it and everything just worked perfectly out of the box. The times where there were compatibility issues with AMD CPU/mobo combos are really over if you get a Nforce 2 board 🙂



[Edit] Had a MSI K7T Pro2a before with 256MB RAM and a TBird 700, and I have to say this baby is faaaaaaaaaaaaast 😀 [/Edit]

 

[RanDum72]

Just one stick of RAM? You aren't even using the full potential of it. The nforce's TwinBank memory architecture runs best when using two sticks of RAM.

 

[ndee]

Originally posted by: RanDum72
Just one stick of RAM? You aren't even using the full potential of it. The nforce's TwinBank memory architecture runs best when using two sticks of RAM.

the other 512MB stick comes in Feburary 🙂 How much faster would it be?

 

[Utterman]

Originally posted by: RanDum72
Just one stick of RAM? You aren't even using the full potential of it. The nforce's TwinBank memory architecture runs best when using two sticks of RAM.

The extra bandwidth from twinbank is for the intergrated video AFAIK. There maybe an improvment if there is no intergrated video, but I'm not totally sure.

 

[ndee]

Originally posted by: Utterman

Originally posted by: RanDum72
Just one stick of RAM? You aren't even using the full potential of it. The nforce's TwinBank memory architecture runs best when using two sticks of RAM.

The extra bandwidth from twinbank is for the intergrated video AFAIK. There maybe an improvment if there is no intergrated video, but I'm not totally sure.

In the Anandtech Nforce2 article, they say it only helps when you use the integrated graphics 🙂

 

[NesuD]

In the Anandtech Nforce2 article, they say it only helps when you use the integrated graphics

Don't know where you saw that but you are mistaken. In the latest article 6-Way Nforce2 Shootout the benchmarks are all run using the dual channel memory configuration for the Nforce2 boards and compared to a pait of kt400 boards that don't have a dual channel memory controller. the advantage of the Nforce 2 boards is plain as day. All the Nforce 2 boards in that article are spp boards (non graphics). There were no IGP boards (integrated graphics) tested in that review. Here is a link to a page in Anands Nforce2 preview part2 that clearly states that is no longer true on the nforce 2 platform nForce2 Part II: Diving Deeper

In Part I we ran the vast majority of our benchmarks with a single stick of memory, utilizing only one of the two 64-bit memory controllers the nForce2 IGP/SPP is equipped with. Our reasoning was that the added bandwidth provided by a 128-bit memory interface is not utilized unless integrated graphics is enabled; this turned out to be both true and false.

the benefit from going to 128-bit DDR mode (DualDDR) is under 3% - less than the normal variance in these benchmarks. Unlike the original nForce however, there are some exceptions to the rule with nForce2.

view the articles benchmark graphs to see exactly where the benefits are at.

 

[BlueWeasel]

Grats on the new system!

Its good to hear that it was a painless installation for you. I am probably about to order the Chaintech Nforce2 board, the XP 2000+, and 2 sticks of the PC3200 256mb RAM in the next few days...

 

[ndee]

Originally posted by: NesuD

In the Anandtech Nforce2 article, they say it only helps when you use the integrated graphics

Don't know where you saw that but you are mistaken. In the latest article 6-Way Nforce2 Shootout the benchmarks are all run using the dual channel memory configuration for the Nforce2 boards and compared to a pait of kt400 boards that don't have a dual channel memory controller. the advantage of the Nforce 2 boards is plain as day. All the Nforce 2 boards in that article are spp boards (non graphics). There were no IGP boards (integrated graphics) tested in that review. Here is a link to a page in Anands Nforce2 preview part2 that clearly states that is no longer true on the nforce 2 platform nForce2 Part II: Diving Deeper

In Part I we ran the vast majority of our benchmarks with a single stick of memory, utilizing only one of the two 64-bit memory controllers the nForce2 IGP/SPP is equipped with. Our reasoning was that the added bandwidth provided by a 128-bit memory interface is not utilized unless integrated graphics is enabled; this turned out to be both true and false.

the benefit from going to 128-bit DDR mode (DualDDR) is under 3% - less than the normal variance in these benchmarks. Unlike the original nForce however, there are some exceptions to the rule with nForce2.

view the articles benchmark graphs to see exactly where the benefits are at.

from http://www.anandtech.com/showdoc.html?i=1654&p=3:

"We?ve proved in the past that the dual channel memory architecture, which NVIDIA is now calling DualDDR, only provides a performance improvement in 3D games with integrated graphics enabled. The reason that adding more memory bandwidth doesn?t improve overall system performance is because the Athlon XP?s FSB is stuck at 133MHz offering up to 2.1GB/s of bandwidth between the Athlon XP and the nForce2 IGP/SPP. With a maximum of 2.1GB/s of data going between the CPU and the IGP/SPP, having twice or even three times that bandwidth between the IGP/SPP and main memory is useless without a memory bandwidth hungry device eating up the remaining bandwidth."






[Edit] OK, read the nforce 2 divin' deeper article. My bad 🙂 [/Edit]

 

[chizow]

ndee is correct.

The AT article is a bad gauge of twinbank b/c it compares nforce2 boards using twinbank vs. KT400. The gains you are seeing are b/c of the overall improvement of nforce2 vs. VIA boards. Tom's (yah I hate to use them as reference, but it applies here) compares nforce2 boards with twinbank vs. single bank and there is virtually no improvement.

Further, sites have speculated the reasoning is the 266mhz Athlon bus can't use the extra bandwidth provided by twinbank as ndee cited above. There is some slight improvement when using the 333mhz Athlon's (only available on the higher XPs) and DDR400, but this starts getting into the pricy range for AMD users. General thoughts are twinbank won't reach its potential until Barton with a full 400mhz bus running DDR400 and 512kb L2 cache, and even then, will only show improvements in certain apps.

Chiz

 

[RanDum72]

Tom's (yah I hate to use them as reference, but it applies here) compares nforce2 boards with twinbank vs. single bank and there is virtually no improvement

Well, there IS an improvement. Check out Tom's charts here.. Some people may have unrealistic expectations (twin memory banks = double performance? NOT!!) but on the lower resolutions under Quake, there is around a 10fps improvement. On other tests, there are small improvements. Not big but there are improvements nonetheless. So why not run the mobo with two sticks of memory?

 

[ndee]

Originally posted by: chizow
ndee is correct.

The AT article is a bad gauge of twinbank b/c it compares nforce2 boards using twinbank vs. KT400. The gains you are seeing are b/c of the overall improvement of nforce2 vs. VIA boards. Tom's (yah I hate to use them as reference, but it applies here) compares nforce2 boards with twinbank vs. single bank and there is virtually no improvement.

Further, sites have speculated the reasoning is the 266mhz Athlon bus can't use the extra bandwidth provided by twinbank as ndee cited above. There is some slight improvement when using the 333mhz Athlon's (only available on the higher XPs) and DDR400, but this starts getting into the pricy range for AMD users. General thoughts are twinbank won't reach its potential until Barton with a full 400mhz bus running DDR400 and 512kb L2 cache, and even then, will only show improvements in certain apps.

Chiz

so I was still right in the end? 🙂 It's kinda..... confusing 🙂 Ya think the Barton will be compatible with the nforce 2 chipset?

 

[ndee]

Originally posted by: RanDum72
Tom's (yah I hate to use them as reference, but it applies here) compares nforce2 boards with twinbank vs. single bank and there is virtually no improvement

So why not run the mobo with two sticks of memory?

Money 😛 I thought long-term on this one. Gonna get the extra stick in a month or two 🙂

 

[BD231]

Dual channel is of no use to the Athlon in any way, even at higher front side bus speeds, I wouldn't imagine it'll do anything for Barton either. Comparing the N-Force 2 and it's dual channel feature to a differet chipset entirly is garbage.

 

[chizow]

Originally posted by: RanDum72
Tom's (yah I hate to use them as reference, but it applies here) compares nforce2 boards with twinbank vs. single bank and there is virtually no improvement

Well, there IS an improvement. Check out Tom's charts here.. Some people may have unrealistic expectations (twin memory banks = double performance? NOT!!) but on the lower resolutions under Quake, there is around a 10fps improvement. On other tests, there are small improvements. Not big but there are improvements nonetheless. So why not run the mobo with two sticks of memory?

I may not be reading those charts correctly, but I think that difference is due to DDR400 vs DDR333. That looks like a mobo shoot-out, the review I was talking about does a comparison of 1 DIMM vs. 2 DIMM on the same board. Regardless, I think its more unrealistic to expect anyone looking at this board to buy it for 10fps at 640x480x16 bit. It would be easier to overclock your video card 2 mhz.

Chiz

Edit: Yes, nForce2 is confirmed to support Barton with bus speeds of 400mhz, allowing users to sync their FSB at 400mhz. Right now the bottleneck is the CPU bus of 266mhz, which provides little or no improvement when paired with DDR 400mhz b/c of the latency/overhead of the DDR400. This is why people are speculating improvement gains may be scene with twinbank b/c the 400mhz synched FSB may be able to use the extra bandwidth provided. But again, I wouldn't expect any miracles (just like HT doesn't do a whole helluva lot).

 

[RanDum72]

That looks like a mobo shoot-out, the review I was talking about does a comparison of 1 DIMM vs. 2 DIMM o

If you look carefully, they have scores for 2-channel and single channel ( 'S' channel) but only for the Asus board, which is perfectly relevant since its the same board the original poster wants to get.

 

[ndee]

Originally posted by: RanDum72
That looks like a mobo shoot-out, the review I was talking about does a comparison of 1 DIMM vs. 2 DIMM o

If you look carefully, they have scores for 2-channel and single channel ( 'S' channel) but only for the Asus board, which is perfectly relevant since its the same board the original poster wants to get.

true, it is a little bit faster. So what else settings can I do to make my computer a little bit faster? 🙂