An exclusive gaming industry community targeted
to, and designed for Professionals, Businesses
and Students in the sectors and industries
of Gaming, New Media and the Web, all closely
related with it's Business and Industry.
A Rich content driven service including articles,
contributed discussion, news, reviews, networking, downloads,
and debate.
We strive to cater for cultural influencers,
technology decision makers, early adopters and business leaders in the gaming industry.
A medium to share your or contribute your ideas,
experiences, questions and point of view or network
with other colleagues here at iVirtua Community.
That one link is your argument as to the advantages of RAID-0? First of all, like I said, a higherr sustained transfer rate does not translate into better performance in most situations. Only where fast data transfer on large blocks of data I/O is important.
I have provided you with plenty of information as to why RAID-0, in most situations, does not offer a performance advantage. Do yourself a favor and actually read what I posted.
Quote:
It proves that I'm very knowledgeable about computers and that my knowledge surpasses yours.
Wow, your joking, right?
Quote:
It looks like it from the way many here are saying how the x1800xt is beating 7800GTX in SLI.
Only one person said that. If you would actually read what I have said in this topic you would clearly see I said that saying nVidia is better then ATI (or vice versa) is just plain ignorant.
Quote:
I guess you don't even want to consider resolutions, power used, and heat?
Power consumption and heat output are things to consider for some, but we are talking about just performance here. Also, yes resolutions do play a factor in the results, but look at any of the benchmarks and what I said is true. Are you really arguing against the point I made that ATI performs better in D3D and nVidia performs better in OpenGL?
Quote:
Well, if it isn't written in the thread, then I can't see it...
PCGEEK said \"I dont see ATI dominating in D3D. Please show me your proof.\" Predator also linked you to the deleted ones.
Quote:
It doesn't look like you go there very much. You certainly didn't show up to debate the RAID 0 issue and it looks as if you don't even know me, so it doesn't look like you have much to go on.
I admit I do not have much to go on, but from the posts I have seen it certaintly looks like the majority of people do not know what they are talking about there. The RAID-0 topic just amazed me.
Quote:
There are some misinformed posts, but saying that an entire forum is filled with uneducated posts is uncalled for. If you can't back that up, then don't say it.
Well, some forums have a lower \"signal to noise ratio\" then others.
Last edited by KoolDrew on Tue Oct 11, 2005 10:39 pm; edited 1 time in total
\"QUOTE It proves that I'm very knowledgeable about computers and that my knowledge surpasses yours.
Wow, your joking, right?\"
I never joke about my intelligence and what I know about computers. I'm not some punk kid that has learned a few things from a web site. I will soon have a Bachelor's degree in Computer Information Systems. That doesn't mean that I know it all, but obviously the skills I have proof of count for a lot. All I've seen out of you is some talk and a couple of links.
Quite frankly, most of the other links that I looked at while browsing for benchmarks showed that, while the performance gain was not huge, it was still there and there was definitely no loss in performance.
Anyway, this argument is going nowhere. You should have stepped up to the plate a while back. Look at benchmark results and time comparisons, and you'll easily see that RAID offers more performance over a single drive. In many of the arguments presented, it was said that it would be best to have the page file on another drive. That means that optimal performance would come from having the page file on a single drive and having the system set up in a RAID 0 array.
I can tell you first hand that my PC loads Windows faster with a RAID 0 system array. If you think otherwise, then you are doing something wrong or have a slower PC where you don't notice the difference...
I never joke about my intelligence and what I know about computers. I'm not some punk kid that has learned a few things from a web site. I will soon have a Bachelor's degree in Computer Information Systems. That doesn't mean that I know it all, but obviously the skills I have proof of count for a lot.
The fact that you said your knowledge surpasses mine when you know nothing about me makes you no better then PCGEEK.
Quote:
All I've seen out of you is some talk and a couple of links.
At least I backed up the facts I was stating. You, on the other hand, are not.
Quote:
Quite frankly, most of the other links that I looked at while browsing for benchmarks showed that, while the performance gain was not huge, it was still there and there was definitely no loss in performance.
Do you mind actually posting them?
Quote:
Look at benchmark results and time comparisons, and you'll easily see that RAID offers more performance over a single drive
I just showed you three articles where RAID-0 did not offer a performance advantage in most situations. So, don't tell me to look at benchmarks.
Rather then just saying I am wrong, actually back up what you say and actually why I am wrong. I have already explained why RAID-0 offers no performance advantage in most situations and I have provided other sources. I highly suggest checking out the StorageReview article and you may actually learn something.
Quote:
In many of the arguments presented, it was said that it would be best to have the page file on another drive. That means that optimal performance would come from having the page file on a single drive and having the system set up in a RAID 0 array.
This is true, beause you can acess the data on both drives concurrently.
Quote:
I can tell you first hand that my PC loads Windows faster with a RAID 0 system array. If you think otherwise, then you are doing something wrong or have a slower PC where you don't notice the difference...
Yes, RAID-0 offers offers a noteable decrease in boot time, but on modern systems boot time is not even close to slow and most people leave their computers on 24/7. You can also use Hibernation instead of doing a complete reboot.
If you want faster boot times for a huge decrease in reliability, sure, go ahead. It just seems really dumb to me.
Quote:
I do lots of file encoding/decoding and video work.
You do not fit under the category of which I am saying will not benefit from RAID-0. The majority of the people who use RAID-0 are gamers, and they will not benefit from it.
Great article. So, there is no need for RAID 0 on a desktop computer. Well, I still think there is a need.
No matter what, my experience with RAID 0 is that you will always get a performance increase. It may be 1% increase or 50% incease, but no matter what, it's an increase. Now depending on your pocket book, one good drive will do the job. But two in RAID 0 won't hurt a bit ;)
Last edited by Super XP on Wed Oct 12, 2005 6:37 pm; edited 1 time in total
With the worsened seek times while using RAID-0 it can decrease performance in certain situations. Especially when it comes to the pagefile because Windows NEVER reads or writes more than 64 KBytes per buffer to the pagefile and they are almost never in sequential 64KB chunks.
You can even check the third link I provided that shows the worsened seek time and load times in some games. Seek time is the dominating factor in most situations, not sustained transfer rate.
Quote:
But two in RAID 0 won't hurt a bit
Like I said, it can hurt and it significantly reduces reliability. Why use something if you will not benefit from it and will significantly increase the risk of losing all your data?
Show me 1 article that says that RAID 0 DECREASES performance more so than increases it, and I might start to believe it... I know that Read speeds are what matters most when running applications that require querying data from the hard drive. If the read speeds are increased, then so is the speed in which the application moves. It's just common sense that pulling data from the hard drive at a faster rate will get it to the processor faster. Even if performance isn't increased for EVERY application used, NO performance is lost. When you DO happen to use an application that benefits from faster Read speeds, then it is worth it.
As far as reliability, who cares? Buy 1 250 GB hard drive. When it goes bad, you lose the entire drive. Lose 2 74GB hard drives in a RAId, and the loss is no greater. Most people don't use more than 2 hard drives, so your complaint of reliability really has no stability. If reliability is an issue, most motherboards allow for at least 4 hard drives. 1 or 2 large storage devices will give you plenty of reliability while a system drive set up as a RAID 0 array will speed up most applications and won't lose any speed when it doesn't.
There is no way that you can convince me that pulling 120GB/s is less performance than a single drive pulling 70GB/s. It doesn't take a math genius to see which is faster.
To help you to understand what is going on, they are referring to the pagefile being on the same drive. Obviously the pagefile is located in a different location than the program that it is also reading data about. Both of these are being queried at almost the same time because both of them are being called on at the same time. That basically means that the head will be running back and forth between the 2 locations, thus slowing everything down because it has to search for the location every time.
With the page file on another drive, this problem is resolved.
Blue, don't even bother arguing with me until you actually read my original post and the links I have provided. Everything you just said I already covered and I am not repeating myself. I even provided links from two of the best review sites there is: Anandtech & StorageReview. Read both of these before you even reply again.
You are completely forgetting that, in most situations, seek time is MUCH more important then sequential transfer rate. This is because the heads are hopping all over the place with IOs to many other files. So, in between all such IOs the heads will have to move anyway. Thge actual time spent reading or writing to the disks platter is small compared to the time seeking or writing to its own buffer.
As for the reliability thing, that also shows you didn't read my first post. I already explained this.
Stop being so close-minded and thinking you are right. Do some reading and you could actually learn something. I have provided you with a pretty long explanation and a few links. If you are not even going to read those don't even bother replieing to this thread.
Quote:
To help you to understand what is going on, they are referring to the pagefile being on the same drive. Obviously the pagefile is located in a different location than the program that it is also reading data about. Both of these are being queried at almost the same time because both of them are being called on at the same time. That basically means that the head will be running back and forth between the 2 locations, thus slowing everything down because it has to search for the location every time.
With the page file on another drive, this problem is resolved.
I already said this...
Last edited by KoolDrew on Wed Oct 12, 2005 7:10 pm; edited 1 time in total
\"Blue, don't even bother arguing with me until you actually read my original post and the links I have provided. Everything you just said I already covered and I am not repeating myself. I even provided links from two of the best review sites there is: Anandtech & StorageReview. Read both of these before you even reply again.\"
First of all, I read that review at Anandtech 6 months ago and I definitely recall calling the editor an idiot that has never used RAID 0. It was quite obvious then, and it is quite obvious now. As far as the Storage review, that was the first one that I read yesterday before replying. I think this must be a case of you not understanding what you are reading becauase I based my argument from the Storage article... The only thing that I've seen as an even valid argument is the page file topic, and I have already covered this twice...
\"You are completely forgetting that, in most situations, seek time is MUCH more important then sequential transfer rate. This is because the heads are hopping all over the place with IOs to many other files. So, in between all such IOs the heads will have to move anyway. Thge actual time spent reading or writing to the disks platter is small compared to the time seeking or writing to its own buffer.\"
It's funny that you would say that I'm forgetting it when I just explained it in my previous post. You would likely learn something if you at least READ my posts instead of assuming that you know it all... You are most definitely wrong, and I would highly recommend that you re-read that Storage article and also do some research about RAID and how it works.
\"As for the reliability thing, that also shows you didn't read my first post. I already explained this.\"
If you have explained this, then why are you stating it like it is of major importance? Obviously it is not.
\"Stop being so close-minded and thinking you are right.\"
lol I don't think I'm right. I know I'm right. Articles are backing me up, as well as benchmarks. You saying that seek times matter more than sequential read times only matters when there are mutliple things that need to be read at the same time that aren't stored in RAM. When you take the Page File out of the equation, that is rare. :)
\"Do some reading and you could actually learn something. I have provided you with a pretty long explanation and a few links. If you are not even going to read those don't even bother replieing to this thread.\"
That's good advice. You should take it. I've made over 2,000 posts while you've made about 300 in your forum? Where's your degree? Where are your builds? Where are your web sites? It looks like someone is behind on their reading, but I don't think that is me...
\"I already said this...\"
Good, then you realize that is the only time when RAID 0 isn't as effective. EVEN THEN, there are still performance gains in some applications. Go back and look at the benchmarks that I posted. Your only argument about it is null and void when taken out of the equation. If it's taken out, then you have no grounds to say that there are no performance gains. You even stated earlier that video editing and encoding offered gains. So, that doesn't count or something? lmao Anyway, you saying that there are no performance gains when there obviously are just doesn't make sense.
First of all, I read that review at Anandtech 6 months ago and I definitely recall calling the editor an idiot that has never used RAID 0.
If he did benchmarks of RAID-0 how the heck could he have not used it? Seriously, just listen to your posts.
I am not even going to continue to argue with you. It is obvious you do not understand how RAID-0 works or anything and no matter how many benchmarks I provide or how much I explain things you will never learn.
I already know from your attitude and ignorance that this will just lead into a big flame war, so I am going to stop now. I suggest you actually do some reading because you are definitely wrong. You still are also not reading what I have said and I am not going to repeat myself.
You really have a LOT to learn and it is evident from your posts here. If you want to learn something go to Ars and post there. A lot of people there could even explain it better then I could as there are a lot of knoweldgable members there. There is one person there that really stands out because he knows a lot about Windows VM. His name is DriverGuru on the forums. He has been writing drivers for Windows NT since before it first shipped (October '92 Driver Development Conference). He used to work with David Solomon, and you'll find him credited in the introduction to \"Windows Internals\" for some of the explanations and diagrams. He runs a successful Windows device driver training and consulting business. Since part of that business involves training for Windows source code program participants, you can probably guess that he has access to Windows source code. He also has access to a lot of the developers. There are tons of other people at the forums there that really do know their stuff.
Last edited by KoolDrew on Wed Oct 12, 2005 8:15 pm; edited 1 time in total
This topic has gone completely off-topic so it is being closed. Blue, if you wish to discuss this further either PM me or you could even sign up at Ars so you can get explanations from many other people. You can just PM me the link and I will participitate in the discussion there. Just post asking the benefits of RAID-0 and just post what your opinion on the matter is. They will say the same thing that I have been saying.
In most situations RAID-0 offers no benefit and it significantly reduces reliability. That is the bottom line. Anybody who argues this is simply wrong.
Digg Topic
Tag topic on On del.icio.us
Technorati Search
Post to Slashdot
