No. Your Raptor drive is plenty fast enough already. You will never notice the difference between one Raptor vs. two Raptors in Raid 0. The raid array is for bragging rights only, though that may be important if you've already sprung for the 6800 processor and 4G of RAM.
No. Your Raptor drive is plenty fast enough already. You will never notice the difference between one Raptor vs. two Raptors in Raid 0. The raid array is for bragging rights only, though that may be important if you've already sprung for the 6800 processor and 4G of RAM.
Are you kidding me? RAID 0 makes a huge difference in speed.
Just try a quick test with The Sims or The Sims 2 for instance. You cut your loading times by around 70% easily.
Try doing any HD benchmarking.
It's RAID 1 that you're thinking of. That doesn't provide a speed boost but offers better data protection.
No. Your Raptor drive is plenty fast enough already. You will never notice the difference between one Raptor vs. two Raptors in Raid 0. The raid array is for bragging rights only, though that may be important if you've already sprung for the 6800 processor and 4G of RAM.
Are you kidding me? RAID 0 makes a huge difference in speed.
Just try a quick test with The Sims or The Sims 2 for instance. You cut your loading times by around 70% easily.
Try doing any HD benchmarking.
It's RAID 1 that you're thinking of. That doesn't provide a speed boost but offers better data protection.
I would agree, I have two raptors and they are very fast! I can boot HL2 really fast.
Don't know about Raid 0 but you're definitely wasting your money with 4 GB of RAM unless you plan on installing XP 64 bit edition - anything short of that and you won't be able to use/see more than 3.25 GB, and even that after a boot.ini tweak.
Last I checked Vista wasn't suppose to come out for another few months. By the time it does come out and you get to buy it, RAM prices will most likely have fallen by another $10-20/stick.
But if you wan instant gratification I can see why you would pull the trigger now :-) ...
1) RAID 1 *does* improve read times as the RAID controller will read simultaneously from both drives just like raid 0 and doesnt have to deal with stripe reconstruction. However it is correct that you will not get extra write performance from RAID 1 like you would get from RAID 0. Loading a game like you mentioned that was mostly reads would probably have similar performance improvements between RAID 0 and RAID 1.
The Intel Matrix RAID controller on the XPS 410 definitely supports this perf enhancement for RAID1. See this info on Intel's site:
2) RAID 0 only gives you improvements for long sequential reads or writes. Lots of random reads and writes actually have better performance with standard single-disk over RAID 0. In many cases RAID 0 is not recommended for gaming applications. Most motherboards ship with "HostRAID" controllers that actually use the CPU to do the striping calculations. This article shows that your FPS are up to 5% lower in most games with RAID 0 due to this load:
There are raid controllers with CPU's built-on, but they are $400+ and wont come built on your motherboard. Combine this with the fact that you are twice as likely to lose your system with a drive failure and RAID 0 is not an ideal gaming setup.
3) If you are absolutely set on RAID 0, get the single drive config and buy the Raptors from newegg or another site listed from pricewatch.com. You wont get gouged on them.
My advice is stick with the cheapest drive Dell will offer. Then buy a 74GB single raptor for ~$149 from newegg as your boot drive and use the dell supplied drive as your data drive where speed is not so important. (If you want the 150GB raptor, its only $224.99 on newegg)
Just running disk benchmarks is a terrible way to decide that RAID 0 is faster. There are some games that do lots of reads of HUGE amount of sequential data that can benefit from RAID 0, but if FPS is important to you, go with single drive.
-David
Morpheus Phreak wrote:
Aivas47a wrote:
No. Your Raptor drive is plenty fast enough already. You will never notice the difference between one Raptor vs. two Raptors in Raid 0. The raid array is for bragging rights only, though that may be important if you've already sprung for the 6800 processor and 4G of RAM.
Are you kidding me? RAID 0 makes a huge difference in speed.
Just try a quick test with The Sims or The Sims 2 for instance. You cut your loading times by around 70% easily.
Try doing any HD benchmarking.
It's RAID 1 that you're thinking of. That doesn't provide a speed boost but offers better data protection.
1) RAID 1 *does* improve read times as the RAID controller will read simultaneously from both drives just like raid 0 and doesnt have to deal with stripe reconstruction. However it is correct that you will not get extra write performance from RAID 1 like you would get from RAID 0. Loading a game like you mentioned that was mostly reads would probably have similar performance improvements between RAID 0 and RAID 1.
The Intel Matrix RAID controller on the XPS 410 definitely supports this perf enhancement for RAID1. See this info on Intel's site:
2) RAID 0 only gives you improvements for long sequential reads or writes. Lots of random reads and writes actually have better performance with standard single-disk over RAID 0. In many cases RAID 0 is not recommended for gaming applications. Most motherboards ship with "HostRAID" controllers that actually use the CPU to do the striping calculations. This article shows that your FPS are up to 5% lower in most games with RAID 0 due to this load:
There are raid controllers with CPU's built-on, but they are $400+ and wont come built on your motherboard. Combine this with the fact that you are twice as likely to lose your system with a drive failure and RAID 0 is not an ideal gaming setup.
3) If you are absolutely set on RAID 0, get the single drive config and buy the Raptors from newegg or another site listed from pricewatch.com. You wont get gouged on them.
My advice is stick with the cheapest drive Dell will offer. Then buy a 74GB single raptor for ~$149 from newegg as your boot drive and use the dell supplied drive as your data drive where speed is not so important. (If you want the 150GB raptor, its only $224.99 on newegg)
Just running disk benchmarks is a terrible way to decide that RAID 0 is faster. There are some games that do lots of reads of HUGE amount of sequential data that can benefit from RAID 0, but if FPS is important to you, go with single drive.
-David
1) RAID 1 will make a small difference in read times as opposed to RAID 0. The reason being that when you are using RAID 1 there is higher latency as compared to RAID 0. So yes there is an improvement over single drive, but not as much.
2) RAID 0 has no real performance hit on newer systems for random read/write especially with processor affinity in a dual-core system. The review your cite is only on a single-core system which has no ability to offload tasks. So picture it more this way. You have twice the CPU power hiding in a dual-core system, and usually one of those cores is sitting unused for the most part and is wide open for doing those calculations.
3) Well this one is self-explanatory. You tell him that if he wants RAID to get the Raptors. Yet shortly after that you tell him to get a single raptor. Thats self-contradictory. Also 1 raptor is not as fast as 2 WD Caviar SE16 drives. In fact the SE16 drives come very close to raptor performance levels due to the extra on-board cache.
No. Your Raptor drive is plenty fast enough already. You will never notice the difference between one Raptor vs. two Raptors in Raid 0. The raid array is for bragging rights only, though that may be important if you've already sprung for the 6800 processor and 4G of RAM.
Are you kidding me? RAID 0 makes a huge difference in speed.
Just try a quick test with The Sims or The Sims 2 for instance. You cut your loading times by around 70% easily.
Try doing any HD benchmarking.
It's RAID 1 that you're thinking of. That doesn't provide a speed boost but offers better data protection.
I'm not talking about HD benchmarks I'm talking about overall system performance. According to a PC mag cover story a couple of months ago the actual overall performance improvement of Raid 0 vs no Raid with comparable drives was on the order of 2-3%.
And did you really do a comparative test of Sims2 loading time with all system parameters held constant except for running things once with Raid 0 and once without? That's pretty hardcore. You reformatted your drives and installed your OS and programs twice just to test the effect of Raid 0? Because that's what you need to do to isolate the performance impact.
And even if you did all that and found your loading time was faster what was the FPS impact? Zero would be my prediction.
"The issue with RAID 0 has always been that splitting data across two hard disks inevitably resulted in doubling the chances of data loss via hard disk failure. It is the logical downside to a single striped volume spanning two physical drives that if either disk fails no data is recoverable. The risk rises as more drives are added to the array. Don't let the "R" in RAID mislead -- in a RAID 0 configuration there is no redundancy.
Associated with a higher risk of data loss RAID 0's only attraction remained its perceived faster speed, and faster speeds are always welcome. It had long been the gripe of video editors that standard hard disks weren't fast enough for video. When technology improvements like higher spindle speeds, larger caches, Tagged Command Queuing (TCQ) etc. brought phenomenal speed increases to the storage arena video editors complained that speeds still weren't sufficient for advanced video work and the handling of higher quality footage - like 10 bit video. Video isn't the only application that takes all the speed that's thrown at it and asks for more. Lots of other applications could use more speed from the IDE subsystem. Since storage speed improvements just haven't matched improvements in other areas like CPUs and GPUs -- and disk reading and writing is still the bottleneck in most modern PCs -- SIs, VARs and even home PC users having been flocking to the technological amphetamine that is RAID 0 instead of spending some time learning how they can optimize their hard disk performance.
What made striping even more attractive was the scalability of the technology. In theory throughput keeps getting faster and faster just by adding more drives to an array.
But very little of that is true. RAID 0 does not always make for more speed. In fact striping may not make the blindest bit of difference to the speed of the average home PC!
Reputed technical websites like StorageReview have often commented that gains on RAID 0 vs single hard disk are minimal at best. Test after test by some of the most reputed technical websites have proved that RAID 0 does not significantly improve desktop performance. Not even with the far higher risk of a four disk RAID 0 array. Seriously! Very few home users tend to be aware of all these technical studies and those that do very often pooh-pooh the idea that the RAID configuration they spent a lot of money on is not actually running faster.
If the claim that RAID 0 is not all it's cracked up to be sounds illogical then it's worth taking the time to read the reviews. A search in Google should lead you to them. Except for a few limited high I/O activities like video editing -- and the typical application benchmark -- the speed gains are almost non-existent. For the average PC user RAID 0 is as useful as a rear spoiler on an 800cc car. It looks good, it sounds impressive but it don't do nuffin'."
I'm not talking about HD benchmarks I'm talking about overall system performance. According to a PC mag cover story a couple of months ago the actual overall performance improvement of Raid 0 vs no Raid with comparable drives was on the order of 2-3%.
And did you really do a comparative test of Sims2 loading time with all system parameters held constant except for running things once with Raid 0 and once without? That's pretty hardcore. You reformatted your drives and installed your OS and programs twice just to test the effect of Raid 0? Because that's what you need to do to isolate the performance impact.
And even if you did all that and found your loading time was faster what was the FPS impact? Zero would be my prediction.
Message Edited by Aivas47a on 10-10-2006 10:46 AM
Ok, now you've left the realm of reality and went into fantasy altogether
Of course RAID0 isn't going to make a difference on the number of frames per second in a game. It's not a video card, so where did you get the idea that it was?
Let me explain some basic PC architecture here. The hard drive stores things. The graphics card renders things. The CPU handles all system calculations other than tasks that are offloaded. Any overflow from the GPU, or NIC, etc. gets handled by the CPU.
Now the faster you have data transferring across the various busses the faster things happen.
The more memory bandwidth you have the faster in memory transfers are. The more GPU/memory bandwidth you have the faster your graphics are. The more network bandwidth you have the faster internal/external transfers are.
Now with RAID0 what it does is make use of 2 competely seperate ATA/SCSI/SATA channels, but treats them as if they were joined as one.
That means files are striped between the 2 drives as if they were one. That's why you have twice the chance of data loss. That's also why you get a nice performance boost as well. Instead of only loading 1 chunk at a time you can now load 2.
That means twice the performance. However that performance doesn't matter if it's small files because of the already negligible load times.
In cases where you are dealing with read/write times on large files though there is a difference. The Sims/Sims2 being a perfect case.
I used to work for a major hardware review site so I often was reformatting and re-installing on my machine to test out different configurations. I can tell you right now that what you are saying makes no sense. Sure loading IE is not going to be much faster. Nor is listening to MP3's or navigating the system.
However when you're loading up games like Half-Life 2, The Sims, Call of Duty, Far Cry, Doom3, etc. that all use large file packaging you will notice a drop of half the time or more required to load the files.
So that last review you cited is somewhat correct. For mom or pop who log onto the net and play bingo, or who just listen to music it won't make a difference. For those of us who actually use our machines it makes a huge difference.
That means files are striped between the 2 drives as if they were one. That's why you have twice the chance of data loss. That's also why you get a nice performance boost as well. Instead of only loading 1 chunk at a time you can now load 2.
That means twice the performance. However that performance doesn't matter if it's small files because of the already negligible load times.
In cases where you are dealing with read/write times on large files though there is a difference. The Sims/Sims2 being a perfect case.
...
What you fail to take into account is that the process of breaking files into stripe chunks to write and then reconstructing has a CPU load that is measurable and has been shown to impact FPS where games can use 100% of CPU even on dual-CPU systems. This is made worse when the games do many random reads/writes of small amounts of data where the overhead of striping causes the overall speed to be slower than if you just had a single drive.
There is no question that RAID 0 is nearly twice as fast for long sequential reads and writes. However this is not the normal drive usage pattern for almost all games where FPS is a factor.
Server's that traditionally used RAID always used expensive controllers that had dedicated CPU's because it was a known issue that RAID reconstruction has a measurable impact on CPU performance, even on systems with 8+ CPU's. It has only recently been a fad to try to provide the same technology for desktop machines at a fraction of the cost. The result is that you are getting the RAID technology without the dedicated CPU, without the dedicated cache and without the battery backup, which for RAID 0 eliminates most of the reasons to use it. RAID 5 and RAID 1 are more practical for the desktop RAID solution because the goal is redundancy, not speed.
If you really want RAID 0 to perform and have $400+, get a LSI Logic MegaRAID 300-4XLP or Adaptec 2420SA which have dedicated processors, onboard cache, and a battery backup to take advantage of write-back caching. Do not think the onboard RAID controllers are providing you anywhere near the performance that these proven server-based products provide.
Anyways, we can respectfully disagree from this point forward and it likely makes a small amount of difference for most users. If you want to feel good that RAID 0 is working for you and your personal experience says it is better, then thats a great thing. I just think it's presumptuous to say it is the ideal solution for everyone buying a gaming machine, especially since it greatly increases the liklihood of failure and data loss.
What you fail to take into account is that the process of breaking files into stripe chunks to write and then reconstructing has a CPU load that is measurable and has been shown to impact FPS where games can use 100% of CPU even on dual-CPU systems. This is made worse when the games do many random reads/writes of small amounts of data where the overhead of striping causes the overall speed to be slower than if you just had a single drive.
I don't fail to take any of that into account. However that CPU load is nearly negligible these days. Even on my old Athlon XP 2500+ system I never saw the CPU go over 2-3% when doing file transfers. Believe me on that machine I did a lot of file transferring as that's the machine I used to backup customer data when doing a reinstall.
Figure 2 years of doing that on the system and I'm a resource mizer. I almost always have a TaskManager window open.
dwlovell wrote:
There is no question that RAID 0 is nearly twice as fast for long sequential reads and writes. However this is not the normal drive usage pattern for almost all games where FPS is a factor.
I really beg to differ on this point here. Have you looked at the files directory for any of the games I mentioned? They all have large level files. That's most of your wait in modern first person shooters. Anything over 32+MB benefits from RAID. Some of those levels and several hundred MB's. Especially in Half-Life 2 & Episode 1.
dwlovell wrote:
If you really want RAID 0 to perform and have $400+, get a LSI Logic MegaRAID 300-4XLP or Adaptec 2420SA which have dedicated processors, onboard cache, and a battery backup to take advantage of write-back caching. Do not think the onboard RAID controllers are providing you anywhere near the performance that these proven server-based products provide.
Anyways, we can respectfully disagree from this point forward and it likely makes a small amount of difference for most users. If you want to feel good that RAID 0 is working for you and your personal experience says it is better, then thats a great thing. I just think it's presumptuous to say it is the ideal solution for everyone buying a gaming machine, especially since it greatly increases the liklihood of failure and data loss.
I agree, having a dedicated card will always be better than having an on-board solution when it comes to how many resources it uses. I also agree that most people aren't hardcore gamers who play games like Half-Life 2. Most PC users are business users at work by a large large percentage. So of course they wouldn't see any difference. Also most home users fit the mom & pop scenario. They wouldn't get any usage out of it either.
However for people like me that do heavy amounts of video/photo editing, lots of hardcore gaming, and do lots of file transfers it makes a lot more sense and I can guarantee you that I get far more than a "2-3%" performance boost.
Just try moving over 4GB of data at one push and watch the difference between single drive and RAID0 and you will see the time drop insanely.
SHUT UP! STOP IT WITH THE HUGE POSTS ALLREADY!!!!!!!!
If u look on the motherboard just under the SATA sockets u will see a chip with a code on it something like this: HN82801GDH that is my 82801GR/GH SATA RAID controller the RAID processor this ensures a performance improvement by taking the load off the CPU.
And I recommend RAID 10 this gives best performance and can tolerate 2 HDD failures.
Alright keep yer nickers on people I am upgrading to RAID10 tomorow and I have to go from RAID0 to a single drive in the process so I will giv u benches on RAID0, single dive, then RAID10 (yes ten read about it).
Aivas47a
1.2K Posts
0
October 8th, 2006 22:00
Morpheus Phreak
297 Posts
0
October 9th, 2006 04:00
winnieB
2 Intern
•
1.1K Posts
0
October 9th, 2006 06:00
I would agree, I have two raptors and they are very fast! I can boot HL2 really fast.
GeekyG
381 Posts
0
October 9th, 2006 12:00
kapalua
2 Intern
•
298 Posts
0
October 9th, 2006 17:00
GeekyG
381 Posts
0
October 9th, 2006 18:00
dwlovell
196 Posts
0
October 9th, 2006 18:00
Morpheus Phreak
297 Posts
0
October 9th, 2006 19:00
Aivas47a
1.2K Posts
0
October 10th, 2006 14:00
Message Edited by Aivas47a on 10-10-2006 10:46 AM
Aivas47a
1.2K Posts
0
October 10th, 2006 15:00
Not to overwork this ... everyone has their opinions ... but here is an assessment from devharware.com that sums up my views on Raid 0 (the full text is at www.devhardware.com/c/a/Storage-Devices/RAID-Not-Such-a-Clever-Idea-for-Your-Home-PC/1/) See also Wikipedia and articles cited there under Raid.
Morpheus Phreak
297 Posts
0
October 10th, 2006 15:00
dwlovell
196 Posts
0
October 10th, 2006 16:00
Morpheus Phreak
297 Posts
0
October 10th, 2006 16:00
GoldSolidGold
12 Posts
0
October 10th, 2006 22:00
If u look on the motherboard just under the SATA sockets u will see a chip with a code on it something like this: HN82801GDH that is my 82801GR/GH SATA RAID controller the RAID processor this ensures a performance improvement by taking the load off the CPU.
And I recommend RAID 10 this gives best performance and can tolerate 2 HDD failures.
GoldSolidGold
12 Posts
0
October 11th, 2006 17:00
RAID0 (stripe):
McAfee Antivirus Disabled
Sequential Writing : 53.91 MB/s (Cpu usage : 7%)
Sequential Read : 85.64 MB/s (Cpu usage : 1%)
Buffered Writing : 1268.68 MB/s (Cpu usage : 48%)
Buffered Reading : 1262.94 MB/s (Cpu usage : 46%)
Random Reading : 142 MB/s (Cpu usage : 48%)
MB/s = Megabyts per second
Mb/s = Megabits per second
There are 8 Bits in a Byte
Message Edited by GoldSolidGold on 10-11-2006 09:13 PM