SATA RAID 0: good for faster game load times?

[strAtEdgE]

Hi,

I had been planning on gettting dual SATA drives in a RAID 0 configuration for my new system, until I started reading all the various posts saying that there was a very negligable gain to be had. It's sometimes hard to discern though what people are talking about in regards to the performance, whether they're speaking to the realistic technical person who knows where gains can and cannot be had in disk performance, or the overzealous newbie who expects it to improve their FPS framerate.

I really had one expectation for RAID 0 and I still have yet to hear it dashed directly: how much, in terms of a %, can RAID 0 help improve the load times of FPS games, both at startup and the change of maps?

Thanks,
strAtEdgE

[EnderW]

http://www.anandtech.com/storage/showdoc.aspx?i=2101&p=10

[strAtEdgE]

Thanks for the article, EnderW. It's spot on with the kind of information I am looking for. The only problem is that the article, in my opinion, is pretty poorly done. I mean they take 1 type of drive and a single RAID controller and test it to determine "RAID 0 doesn't work". I would hope it's common sense that some RAID controllers are not as good as others. It would be nice to see something with a little more variety.

But none the less, score a point for not bothering with RAID 0.

[EnderW]

Thanks for the article, EnderW. It's spot on with the kind of information I am looking for. The only problem is that the article, in my opinion, is pretty poorly done. I mean they take 1 type of drive and a single RAID controller and test it to determine "RAID 0 doesn't work". I would hope it's common sense that some RAID controllers are not as good as others. It would be nice to see something with a little more variety.

But none the less, score a point for not bothering with RAID 0.
here are some more articles with a little more information

http://faq.storagereview.com/tiki-index.php?page=SingleDriveVsRaid0
http://www.storagereview.com/articles/200406/20040625TCQ_1.html

also consider this stunning example:

In the SR Gaming DriveMark… a 4 drive Cheetah array.. lags a single Raptor running on the "dumbest" of SATA controllers by a margin of 9%.

[strAtEdgE]

Here's a great article I found on the subject in support of RAID 0 : http://www.tweakers.net/reviews/515

[EnderW]

Here's a great article I found on the subject in support of RAID 0 : http://www.tweakers.net/reviews/515

interesting article, but did you read this part?

Our gaming traces show that most games have very little impact on modern hard drives and generate a rather low load. Even the heavyweight champion Battlefield Vietnam was able to burden a Raptor WD360GD for only 17,1 percent average with a short peak of 70 percent. Other games had even lower averages and didn't rise above 80 percent peak usage. Therefore we conclude that loading game levels is mostly cpu intensive and does not rely on storage devices. This conclusion is again illustrated by the insignificant improvents when loading levels in Far Cry and Unreal Tournament. Also, measurements with Unreal Tournament show some unlikely differences between a Raptor WD740GD and the other disks in the arena. It seems to be 12,9 percent faster than number two, a Seagate Barracuda 7200.7, while the others (one of them is a Raptor 360GD and a rather old IBM Deskstar 75GXP) offer a performance varying not more than 8,4 percent.

[Ice Czar]

Femme Taken over at tweakers.net employs the same basic set of utilities that Storagereview does for his benchmarks
SPT IPEAK (http://www.storagereview.com/articles/200111/20011109Renaissance_5.html)
but has historically done much more RAID Benchmarking than either SR or Anandtech
http://www.hardforum.com/showthread.php?s=&postid=1026104985#post1026104985

Im digesting that info, thanx I hadnt seen that ;)

[strAtEdgE]

interesting article, but did you read this part?

Yeah, I had, and I've accepted the fact that RAID 0 is not going to help for gaming. But now I'm considering whether or not it would help with file sharing, copying, installing, loading the OS, etc. I'm on the fence, I think perhaps I'll start with a single drive and if I find I'm striving for an improvement in disk performance somewhere, I'll pick up a second one.

[shieldforyoureyes]

It will only help with file-sharing if your network connection
is faster than your drives.

[[H]Rabbit]

Unless it's too late - benchmark the single drive configuration with some real world tests (aka stopwatch + progs and bench progs). That way if you decide to try two drives it won't be a matter of them "feeling" faster, slower, or the same - it WILL be faster, slower, or the same.

[djnes]

Yeah, I had, and I've accepted the fact that RAID 0 is not going to help for gaming. But now I'm considering whether or not it would help with file sharing, copying, installing, loading the OS, etc. I'm on the fence, I think perhaps I'll start with a single drive and if I find I'm striving for an improvement in disk performance somewhere, I'll pick up a second one.

It's not going to be faster for any of those uses either. I've done the tests with several sets of hard drives and it doesn't really make a difference.

[trust_no1]

Thanks for the article, EnderW. It's spot on with the kind of information I am looking for. The only problem is that the article, in my opinion, is pretty poorly done. I mean they take 1 type of drive and a single RAID controller and test it to determine "RAID 0 doesn't work". I would hope it's common sense that some RAID controllers are not as good as others. It would be nice to see something with a little more variety.

But none the less, score a point for not bothering with RAID 0.

Consider the fact that the are using the fastest SATA HDD...

[djnes]

Consider the fact that the are using the fastest SATA HDD...

I'm not sure why people are hung up on this fact, but here goes the common sense explanation.

It doesn't matter what drives your using. As long as your single drive test uses one of the drives from the RAID array, it doesn't matter! By using one of the drives, your testing the RAID functionality, and only the RAID functionality. This is something people learn in grade school with the scientific method. They weren't comparing 2 old drives in RAID0 to a new Raptor. This is just plain common sense and logic as far as testing methods go.

It's wrong to say the results are skewed because the single drive test was done using a Raptor...IF the RAID0 array used the same drives.

[[H]Rabbit]

I'm not sure why people are hung up on this fact, but here goes the common sense explanation.

It doesn't matter what drives your using. As long as your single drive test uses one of the drives from the RAID array, it doesn't matter! By using one of the drives, your testing the RAID functionality, and only the RAID functionality. This is something people learn in grade school with the scientific method. They weren't comparing 2 old drives in RAID0 to a new Raptor. This is just plain common sense and logic as far as testing methods go.

It's wrong to say the results are skewed because the single drive test was done using a Raptor...IF the RAID0 array used the same drives.

There is some validity to it if a single drive is already capable of saturating the transfer bus. In that case then of course you wouldn't see an increase in speeds from RAID 0...

Then again modern HDDs don't come close to bus limits - so it doesn't matter if you're using a Raptor or a 4200RPM laptop drive. As long as you compare a drive used in the array to the array (of identical drives) you get a valid indicator of performance for that test, all else being equal, regarding single drive versus RAID performance.