Is VIA bad for raid?

[aarona]

http://www.tecchannel.de/hardware/817/index.html
Has anyone here seen this report or others like it?

 

[girish]

I havent read the article, but I guess it isnt true.

VIA chipsets traditionally had problems with their Southbridges especially the IDE controller that lies in it. IDE RAID is 100% dependendt on the IDE controller that works under the RAID controller, which fortunately isnt made by VIA. The HighPoint or Promise RAID controllers use their own IDE controllers so VIA doesnt really come into the picture.

For the system, the RAID array appears to be a single disk so it isnt concerned with rest of the system especially the Northbridge either.

If you are working with a SCSI RAID controller even then its either Adaptec or AMI whatever controller you use, its just a PCI device for the system.

girish

<font color=red>Nothing is fool-proof. Fools are Ingenious!</font color=red>

 

[Ncogneto]

Dude, you really ought to read the article before passing on assumptions like this. Fact of the matter it is true. If you took the time to read the article you will find that the problem lies within limitations of the pci bus, which, at least to my knowledge, would affect all SCSI cards. even the onboard controllers are still connected via the pci bus.



Via still has a long way to go, and why I still strongly suggest chipsets by others as an alternative.

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

 

[Ncogneto]

The problems found, especially with the frequent interruptions of data transfers on the PCI bus of VIA's chipsets, may also be responsible for many of the compatibility issues found with their products in the past. VIA confirmed our findings shortly after the initial german release. The chipmaker said that he was working on a solution. Whether that will be new drivers, a patch or a complete redesign of the chipsets is unknown so far.

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

 

[Ron_Jeremy]

So far VIA is bad for everything, not just RAID. I think the only thing VIA is good at is hypothetical SiSoft Sandra Memory benchmarks.

Well, hold on for a second, I jumped to conclusions too quickly. I guess you could say VIA is good for keeping this particular forum alive. If it wasn't for VIA boards, this would be a dead forum.

Crashman, where are you?

Cheers,

Ron_Jeremy

If you loan a friend $20 & never see them again, it was worth it.

 

[Kronos]

Odd...my IWILL KK266 with VIA Chipset works flawlessly.

I want to die like my Grandfather...in my sleep...not screaming in terror like his passengers.

 

[mbetea]

ya, same with my ga-7vtx-p. i should just trash the thing because it's working perfeclty and go for another k7s5a, hopefully with any luck i'll be able to send it back at least once for replacement! :lol:

[insert philosophical statement here]

 

[Ncogneto]

You could send that ga-7vtx-p back a thousand times and still have the issues talked about in this article.

It's not what they tell you, its what they don't tell you!

 

[digital_trucker]

I'm in love with my MSI 266a.

Santa must be a trucker.

Dorr, MI

 

[girish]

Well I dint read the article then since I was in a hurry, and I just commented on generic level. Now that I have read it, I must say they have done quite a good job pointing this out.

But in practice many onboard RAID boards just work fine, I dont dare to use a VIA board (I had investigated the Asus CUV4X-D for it) for my servers so I dont know how they screw a SCSI RAID, but for me I havent found anything wrong with the VIA PCI controller. The benchmarks dont say so.

There might be certain PCI timing issues with the VIA PCI controller (and the AGP controller has some too). But then, almost all other hardware on the board works on the PCI bus so the cumulative effect of all this could be seen in the resultant performance which is not.

Perhaps the weekly VIA 4in1 drivers do have some patch that covers the problem, that doesnt appear in everybody else's computer. You would have millions of people arying about the poor performance of the VIA chipsets. But they simply happen to be among the best performing chipsets for all platforms!

girish

<font color=red>Nothing is fool-proof. Fools are Ingenious!</font color=red>

 

[mbetea]

well from the article i'm gathering it's effecting ata133? well with ata100, i have the same speeds as i do on 2 other boards, so i'm not worried. my next rig will be a dual setup and when i get around to buying newer drives, it will be for a new comp. though it is a good article and it clearly shows how via is behind everyone else.

[insert philosophical statement here]