Kinda bogus review here. How on earth can you put a SATA drive in comparison to other SATA drives, then draw conclusions about SATA vs SCSI?? The answer is you can't, especially in regards to SATA being 'ready' for use in big fileservers. It's a cheap alternative to SCSI, not a performance alternative. Check the link below for fileserver iometer (IOPs) benchmarks of many different drives, and you'll find that the top SCSI drives demolish the top SATA drives with nearly double the IOPs ratings :
I've said it before but I'll say it again... Gary sets the standard for technical writing excellence. His work should be studied closely by all who aspire to do this type of work. I hope you're at least 26 or older becasue if you're writing like this at 19, 20 or 21, I think I'm going to get sick... with jealousy.
I will see if we can get a WD4000 series and update the article or at least have it for the RAID follow up. We have the new WD 500GB drive to test also.
Gary - Thanks for this excellent review and for addressing the two primary complaints I had with the last harddrive review here: The old Samsung drives listed and lack of SP2405C and the tiny decibel test graph that was hard to read and draw conclusions from.
This test has a great lineup of drives tested and much better graphs. Very nicely done!
quote: Gary - Thanks for this excellent review and for addressing the two primary complaints I had with the last harddrive review here: The old Samsung drives listed and lack of SP2405C and the tiny decibel test graph that was hard to read and draw conclusions from.
Thank you and actually everyone for the comments. We have certainly listened to the requests and hopefully our improvements over the course of the next few articles will be satisfactory for everyone.
Although I have had some personal issues with Samung drives in the past I have to give them credit for producing a very good drive in the SP2504c. It would be my drive of choice in a HTPC or SFF setup at this time if you need a silent drive. The performance of the drive is competitive with others in its class for the most part at a very good price point. I tested the drive last night in a sound enclosure and there was only a 5 decibel difference between idle and constant read/write operations. I believe some of the additional decibel readings in the article is our equipment picking up noise from the drive cage. We will continue to refine this process for testing but the readings we reported were very consistent given the environmental conditions in the test room.
It would be great if you could include a comparison of the all 3 raptors in Raid 0. I have a 36gb Raptor and have often wondered if 2 in raid 0 would be as fast as 1 74gb raptor. Esp. now w/newer versions.
The content was very informative, and as a whole, the article was very well done. But jesus christ, dude, opening your articles with quotes ticks me off to no end.
Perhaps the quotes at the front of Gary's reviews will give you a repetoire that will include words that can be substituted for Dude. If you hate the quote just skip the first few sentences and you will not have to be bothered by the quotes.
Personally I like them - the quotes set the reviews apart. It's sometimes nice to have a little education with my favorite hobby.
quote: The content was very informative, and as a whole, the article was very well done. But jesus christ, dude, opening your articles with quotes ticks me off to no end.
Is it the quote itself or just the fact there is one? I would appreciate some honest feedback on the subject. I like the quote as it sets the tone for the article but I also realize it really is not needed. :-)
I love the quotes. In particular, this one stuck in my mind, and provided an incredibly simple, single-sentence "hook" into the review which summed it up perfectly. I'd vote to keep them.
newbie q: the review uses the term "single user" recommending turning off native command queing in single user setting as opposed to multi user. i have a pair of wd1500s on my floor ready to be installed in our office system in Raid 0. One person at a time uses the computer but we frequently "multi-ask". do we turn of NCQ to get faster speeds in this sitaution--am I a "single user" on this computer, though much of our use is multi-asking--six programs at once e.g. doing multiple tasks.
Hi,
I am assuming your applications will be general office and the machine is not being setup as a local server. If so, turn off NCQ. It it really meant at the server-level where you have highly random and concurrent requests. I would not go with RAID 0 for an office environment as you are playing with fire in regards to data reliability. Setup you system in RAID 10 if it supports it.
Go to your system props>device manager>SATA controller, and uncheck the option.
I have the latest Nforce 4 drivers loaded, so I cant say its exactly the same for earlier versions, or Intel/other chipsets, but that would be the logical place to look.
This review goes into way too much depth about NCQ vs. no NCQ. I can't remember if they say how to disable it. It is also a great review comparing SCSI vs. Raptor and SATA vs. Raptor performance. It is the most indepth review to date of all raptor drives except it leaves out Raid tests.
quote: Whoops, I meant to imply that the Storagereview.com review goes into way too much detail about NCQ, not the Anandtech review. :)
I understood that from your original statement. We wanted to go into more detail but the new game and multi-tasking benchmarks were not ready yet. Hopefully, we can provide more details in the near future.
quote: This review goes into way too much depth about NCQ vs. no NCQ. I can't remember if they say how to disable it. It is also a great review comparing SCSI vs. Raptor and SATA vs. Raptor performance. It is the most indepth review to date of all raptor drives except it leaves out Raid tests.
Eugene Ra provides excellent information and analysis. :-) We are expanding our test suites at this time and will be providing a more in-depth analysis of drives in the near future. Also, we will be expanding our RAID coverage depending upon the drive being tested (will it make a difference). There is certainly a need to do RAID analysis but it should include more than RAID 0 which only provides benefits in certain benchmarks. ;->
I noticed you revised your acoustic testing, but this leads to a different ranking when compared to the one in the article :
"Seagate 7200.9 160GB: The Highest Platter Density to Date!"
http://anandtech.com/storage/showdoc.aspx?i=2682&a...">http://anandtech.com/storage/showdoc.aspx?i=2682&a... In the article mentionned above, the idle noise of the 500 GB 7200.9 is lower than the idle noise of the 74 GB Raptor, and in this article, it is the contrary.
What is the explanation of this ?
Also, did you test the acoustics of the Samsung SP2504C with the "old" method ? I am interested by the comparison with other models (I try to have the most silent drives).
Id like to see some raided drive scores as well. Because this drive is marketed at enthusists . . . good chance some will just buy two and raid them. Majority of the readers don't have scsi drives, so not sure why you'd choose those benchmarks over some consumer raided drives.
So RAID was debated in the comments of the last HDD article AT posted, and while traditionally RAID hasn't shown much of a performance improvement, the fact that nearly all enthusiast motherboards now come with some sort of RAID controller and since the Raptors now use a native Serial ATA interface, the story might have changed.
Can AT do a RAID shootout or something? I would be very interested in something like that.
quote: Waste of time but I guess when the RAID freaks demand it they get it.
Maybe we should do a RAID 5 test only.....the RAID 0 results are what you expect them to be at this time, very good in two system benchmarks but slower in a couple of games..... ;->
The THG review only compares a single 150 Raptor to a RAID 0 with slower drives, but it does use a few real-world benchmarks. The other review tests RAID 0 with these Raptors but only uses synthetic benchmarks. Synthetic benchmarks are useful, but they are only part of the performance picture. Results with real-world benchmarks - particularly in RAID 0 - are very different.
The tests linked are not very good. GamePC uses synthetic benchmarks which are not indicative of real world desktop performance. The Iometer in particular is not useful for single user scenarios and never was. Basically Gamepc doesn’t know what they are doing (or since they sell the stuff maybe that’s just what they do...)
We all RAID-O improves performance a lot in Synthetic benchmarks. It does however not do the same for real world tests and certainly not in a cost effective way. I highly doubt SATA or this new raptor changes that. But would be interesting nonetheless to see some tests on the matter.
This is the best hard drive review I've seen on AT in a very long time. Great job, guys. The history of Raptors at the start was a nice touch to help me better understand what WD is doing. I also think this is the first drive review I have seen by Gary Key. Bring us some more like this one.
I have to agree this was really a step up as far as hardrive reviews go. Nice to see you are improving your thermal and acoustics testing. Keep up the good work!!!
Agreed - great review, and it was really interesting and useful to get an understanding of the evolution of the product.
My only question is, why wasn't Raptor+NCQ added as part of the multi-tasking tested? I can see that being a useful indicator of desktop performance, especially as dual-core CPUs become more prevalent.
We will be investigating the effects of NCQ in more detail in the near future. We are working on a revised benchmark suite that consists of more real world applications and varying multi-task scenarios.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
51 Comments
Back to Article
shady28 - Sunday, April 9, 2006 - link
Kinda bogus review here. How on earth can you put a SATA drive in comparison to other SATA drives, then draw conclusions about SATA vs SCSI?? The answer is you can't, especially in regards to SATA being 'ready' for use in big fileservers. It's a cheap alternative to SCSI, not a performance alternative. Check the link below for fileserver iometer (IOPs) benchmarks of many different drives, and you'll find that the top SCSI drives demolish the top SATA drives with nearly double the IOPs ratings :
http://www.storagereview.com/comparison.html">http://www.storagereview.com/comparison.html
Kensei - Friday, February 10, 2006 - link
I've said it before but I'll say it again... Gary sets the standard for technical writing excellence. His work should be studied closely by all who aspire to do this type of work. I hope you're at least 26 or older becasue if you're writing like this at 19, 20 or 21, I think I'm going to get sick... with jealousy.Kensei
SpecOps - Friday, February 10, 2006 - link
I would have liked to see the "value Raptor" in the tests as well, so we can get a better picture of price vs. performance.Gary Key - Friday, February 10, 2006 - link
I will see if we can get a WD4000 series and update the article or at least have it for the RAID follow up. We have the new WD 500GB drive to test also.JustAnAverageGuy - Thursday, February 9, 2006 - link
Good article Gary, look forward to more in the future.One of the few editors who actually reads and replies to more than just one or two comments. :thumbsup;
- JaAG
CrystalBay - Wednesday, February 8, 2006 - link
Gary , you have a very proficient writing ability . I noticed it on your first review at AT keep up the good work...MadAd - Wednesday, February 8, 2006 - link
I just checked and prices for these in the UK is £210 - its almost the price of a scsi drive ffs (a 147gig hitachi u320 is only £50 more).
Im not quite sure what market segment these are aimed at, I just hope they dont shoot themselves in the foot being priced so close.
yacoub - Wednesday, February 8, 2006 - link
Gary - Thanks for this excellent review and for addressing the two primary complaints I had with the last harddrive review here: The old Samsung drives listed and lack of SP2405C and the tiny decibel test graph that was hard to read and draw conclusions from.This test has a great lineup of drives tested and much better graphs. Very nicely done!
Gary Key - Thursday, February 9, 2006 - link
Thank you and actually everyone for the comments. We have certainly listened to the requests and hopefully our improvements over the course of the next few articles will be satisfactory for everyone.
Although I have had some personal issues with Samung drives in the past I have to give them credit for producing a very good drive in the SP2504c. It would be my drive of choice in a HTPC or SFF setup at this time if you need a silent drive. The performance of the drive is competitive with others in its class for the most part at a very good price point. I tested the drive last night in a sound enclosure and there was only a 5 decibel difference between idle and constant read/write operations. I believe some of the additional decibel readings in the article is our equipment picking up noise from the drive cage. We will continue to refine this process for testing but the readings we reported were very consistent given the environmental conditions in the test room.
Jeff7181 - Wednesday, February 8, 2006 - link
I agree with those that say a Part 2 with RAID testing would be interesting.Gary Key - Wednesday, February 8, 2006 - link
Believe me, Purav cannot wait to start testing this drive in a RAID setup.feraltoad - Wednesday, February 8, 2006 - link
It would be great if you could include a comparison of the all 3 raptors in Raid 0. I have a 36gb Raptor and have often wondered if 2 in raid 0 would be as fast as 1 74gb raptor. Esp. now w/newer versions.Rapsven - Wednesday, February 8, 2006 - link
The content was very informative, and as a whole, the article was very well done. But jesus christ, dude, opening your articles with quotes ticks me off to no end.rjm55 - Wednesday, February 8, 2006 - link
Perhaps the quotes at the front of Gary's reviews will give you a repetoire that will include words that can be substituted for Dude. If you hate the quote just skip the first few sentences and you will not have to be bothered by the quotes.Personally I like them - the quotes set the reviews apart. It's sometimes nice to have a little education with my favorite hobby.
Gary Key - Wednesday, February 8, 2006 - link
Is it the quote itself or just the fact there is one? I would appreciate some honest feedback on the subject. I like the quote as it sets the tone for the article but I also realize it really is not needed. :-)
Rolphus - Thursday, February 9, 2006 - link
I love the quotes. In particular, this one stuck in my mind, and provided an incredibly simple, single-sentence "hook" into the review which summed it up perfectly. I'd vote to keep them.johnsonx - Wednesday, February 8, 2006 - link
I like the quotes. It makes me think of Chairman Kaga on Iron Chef (the original Japanese show, not the new Iron Chef America).Seriously Gary, keep the quotes.
fb0252 - Wednesday, February 8, 2006 - link
newbie q: the review uses the term "single user" recommending turning off native command queing in single user setting as opposed to multi user. i have a pair of wd1500s on my floor ready to be installed in our office system in Raid 0. One person at a time uses the computer but we frequently "multi-ask". do we turn of NCQ to get faster speeds in this sitaution--am I a "single user" on this computer, though much of our use is multi-asking--six programs at once e.g. doing multiple tasks.Gary Key - Wednesday, February 8, 2006 - link
Hi,I am assuming your applications will be general office and the machine is not being setup as a local server. If so, turn off NCQ. It it really meant at the server-level where you have highly random and concurrent requests. I would not go with RAID 0 for an office environment as you are playing with fire in regards to data reliability. Setup you system in RAID 10 if it supports it.
Thanks....
Gary Key - Wednesday, February 8, 2006 - link
We need an edit function. "Setup your system in RAID 10 if it supports it."retrospooty - Wednesday, February 8, 2006 - link
How do you enable/disable NCQ? I can't find anything on that.Gary Key - Wednesday, February 8, 2006 - link
This is accomplished through the NVIDIA SW-IDE driver settings. I will post a how to this evening.
retrospooty - Wednesday, February 8, 2006 - link
Thanks, I found it, based on your clue above... I was looking at the drive props, not the IDE controller props... Duh.Live - Wednesday, February 8, 2006 - link
Care to enlighten the rest of us?
retrospooty - Wednesday, February 8, 2006 - link
Go to your system props>device manager>SATA controller, and uncheck the option.I have the latest Nforce 4 drivers loaded, so I cant say its exactly the same for earlier versions, or Intel/other chipsets, but that would be the logical place to look.
Live - Wednesday, February 8, 2006 - link
I am wondering the same thing. Since it seems NCQ is not worth it for desktop use how do you disable it?mlittl3 - Wednesday, February 8, 2006 - link
This review goes into way too much depth about NCQ vs. no NCQ. I can't remember if they say how to disable it. It is also a great review comparing SCSI vs. Raptor and SATA vs. Raptor performance. It is the most indepth review to date of all raptor drives except it leaves out Raid tests.http://www.storagereview.com/articles/200601/WD150...">http://www.storagereview.com/articles/200601/WD150...
mlittl3 - Wednesday, February 8, 2006 - link
Whoops, I meant to imply that the Storagereview.com review goes into way too much detail about NCQ, not the Anandtech review. :)Gary Key - Wednesday, February 8, 2006 - link
I understood that from your original statement. We wanted to go into more detail but the new game and multi-tasking benchmarks were not ready yet. Hopefully, we can provide more details in the near future.
Gary Key - Wednesday, February 8, 2006 - link
Eugene Ra provides excellent information and analysis. :-) We are expanding our test suites at this time and will be providing a more in-depth analysis of drives in the near future. Also, we will be expanding our RAID coverage depending upon the drive being tested (will it make a difference). There is certainly a need to do RAID analysis but it should include more than RAID 0 which only provides benefits in certain benchmarks. ;->
jamescleant - Wednesday, February 8, 2006 - link
I noticed you revised your acoustic testing, but this leads to a different ranking when compared to the one in the article :"Seagate 7200.9 160GB: The Highest Platter Density to Date!"
http://anandtech.com/storage/showdoc.aspx?i=2682&a...">http://anandtech.com/storage/showdoc.aspx?i=2682&a...
In the article mentionned above, the idle noise of the 500 GB 7200.9 is lower than the idle noise of the 74 GB Raptor, and in this article, it is the contrary.
What is the explanation of this ?
Also, did you test the acoustics of the Samsung SP2504C with the "old" method ? I am interested by the comparison with other models (I try to have the most silent drives).
Gary Key - Wednesday, February 8, 2006 - link
I will detail a response tonight and test the SP2504c with the old standard.Gary Key - Wednesday, February 8, 2006 - link
I am still working on providing the old numbers. I did locate the original sound meter used and will have it here tomorrow.noxipoo - Wednesday, February 8, 2006 - link
3 seagate 7200.9 in raid 0 vs 1 raptor? about the same price point, would be interesting.mlittl3 - Wednesday, February 8, 2006 - link
See my post above. Gamepc did 4 rapters in raid 0 versus scsi and P-ATA drive.sxpose - Wednesday, February 8, 2006 - link
Id like to see some raided drive scores as well. Because this drive is marketed at enthusists . . . good chance some will just buy two and raid them. Majority of the readers don't have scsi drives, so not sure why you'd choose those benchmarks over some consumer raided drives.mlittl3 - Wednesday, February 8, 2006 - link
See my post above for raid tests.Orbs - Wednesday, February 8, 2006 - link
First of all, another great article, AT!So RAID was debated in the comments of the last HDD article AT posted, and while traditionally RAID hasn't shown much of a performance improvement, the fact that nearly all enthusiast motherboards now come with some sort of RAID controller and since the Raptors now use a native Serial ATA interface, the story might have changed.
Can AT do a RAID shootout or something? I would be very interested in something like that.
Again, great article!
Gary Key - Wednesday, February 8, 2006 - link
We will have updated benchmarks or a new article with RAID results for this drive and others in the near future. Thank you.Zebo - Saturday, February 11, 2006 - link
Waste of time but I guess when the RAID freaks demand it they get it.Gary Key - Monday, February 20, 2006 - link
Maybe we should do a RAID 5 test only.....the RAID 0 results are what you expect them to be at this time, very good in two system benchmarks but slower in a couple of games..... ;->
mlittl3 - Wednesday, February 8, 2006 - link
Here you go.http://www.gamepc.com/labs/view_content.asp?id=rap...">http://www.gamepc.com/labs/view_content.asp?id=rap...
http://www.tomshardware.com/2006/02/06/wd1500ad_ra...">http://www.tomshardware.com/2006/02/06/wd1500ad_ra...
Two articles with raid numbers.
Wesley Fink - Wednesday, February 8, 2006 - link
The THG review only compares a single 150 Raptor to a RAID 0 with slower drives, but it does use a few real-world benchmarks. The other review tests RAID 0 with these Raptors but only uses synthetic benchmarks. Synthetic benchmarks are useful, but they are only part of the performance picture. Results with real-world benchmarks - particularly in RAID 0 - are very different.Live - Wednesday, February 8, 2006 - link
The tests linked are not very good. GamePC uses synthetic benchmarks which are not indicative of real world desktop performance. The Iometer in particular is not useful for single user scenarios and never was. Basically Gamepc doesn’t know what they are doing (or since they sell the stuff maybe that’s just what they do...)We all RAID-O improves performance a lot in Synthetic benchmarks. It does however not do the same for real world tests and certainly not in a cost effective way. I highly doubt SATA or this new raptor changes that. But would be interesting nonetheless to see some tests on the matter.
rjm55 - Wednesday, February 8, 2006 - link
This is the best hard drive review I've seen on AT in a very long time. Great job, guys. The history of Raptors at the start was a nice touch to help me better understand what WD is doing. I also think this is the first drive review I have seen by Gary Key. Bring us some more like this one.Live - Wednesday, February 8, 2006 - link
I have to agree this was really a step up as far as hardrive reviews go. Nice to see you are improving your thermal and acoustics testing. Keep up the good work!!!Rolphus - Wednesday, February 8, 2006 - link
Agreed - great review, and it was really interesting and useful to get an understanding of the evolution of the product.My only question is, why wasn't Raptor+NCQ added as part of the multi-tasking tested? I can see that being a useful indicator of desktop performance, especially as dual-core CPUs become more prevalent.
Rolphus - Wednesday, February 8, 2006 - link
That would be multi-tasking tests. Only had one coffee so far.Gary Key - Wednesday, February 8, 2006 - link
We will be investigating the effects of NCQ in more detail in the near future. We are working on a revised benchmark suite that consists of more real world applications and varying multi-task scenarios.ohnnyj - Wednesday, February 8, 2006 - link
In some of the graphs there are two 74GB Raptors, are these two different revisions?John
Gary Key - Wednesday, February 8, 2006 - link
Yes, the WD740GD-00FLC0 is the drive listed under the WD740GD description. I will have an additional notation up shortly.