What's new
  • Please do not post any links until you have 3 posts as they will automatically be rejected to prevent SPAM. Many words are also blocked due to being used in SPAM Messages. Thanks!

Intel X-25M 80GB SSD Review

Status
Not open for further replies.

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,274
intelcorplogo.jpg


Intel X-25M 80GB SSD Review




Manufacture Page: Intel® Mainstream SATA Solid State Drives
Part Number: SSDSA2MH080G1C5
TechWiki Info: Intel X-25M SSD
Price: Click Here to Compare Prices
Warranty: 3 year




It certainly has been a while since we last looked at the state of Solid State drives, but this doesn’t mean we have not been keeping an eye on things for you, our loyal readers. Like a Master saucier who sips and sips and only when the stew is truly ready allows it to be released, we too have been biding our time waiting for the market situation to be just right before entering back into the SSD review arena. In the past we have looked at SLC based Solid State drives but we never did look at the first gen of MLC based drives and we intend to continue to ignore them as they did have major growing pains. Now that the second generation MLCs are out and 3rd gens are just starting to appear, we think its time to go back to the review board, update our tests and take another look at the state of SSDs.

Today we are going to start off our re-entry into SSD reviews by looking closely at the Intel X-25M. This well-known drive comes in a moderate size of 80GB and also 160GB. Our drive is the 80GB model but the differences in speed and performance between the two models should be minor; however, if you are looking for a cheap SSD, neither is what you would call affordable. The 80GB drive has been on the market for a little while so it is widely available from retailers and e-tailers alike, but this version goes for about $450 to $500 or $5.62 per GB for those of you keeping score at home.

Intel as a company needs no introduction as they are the 800lb. gorilla when it comes to CPUs; however this is their first real foray into “mainstream” SSDs. As such it should be interesting to see how effective the first generation is, and whether or not it suffers from “first version syndrome”. We truly doubt that it will considering if anything, Intel is known for their approach to solid QA so while it may not end up being a speed demon we highly doubt many in the near future will beat it in the overall performance category.

As such, this review is going to focus not only on drive performance, but also determine if this product will be our benchmark high-water mark for future reviews. In any case this is going to be blast; so sit back, relax and be prepared to be amazed at what an MLC based drive can do.

pic.jpg

 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,274
Specifications

Specifications


specs.jpg

 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,274
Packaging and Accessories / First Impressions

Packaging and Accessories


box_sm.jpg
box_open_sm.jpg

Where this unit is not a Retail Market drive (it’s actually an engineering sample) we are going to skip the external packaging and accessories parts of this review as it would not be fair to Intel to do so. As you can see our X-25M came in a cardboard box with foam internal packaging but yours may not or may not come like this.


First Impressions


top_sm.jpg

Moving on to the X-25M itself, you can see it is done is finished in a very unique black, silver and white colour scheme much like Intel's Extreme Edition processor boxes. Honestly, if you didn’t know what kind of power was lurking inside the plain cardboard box, you would have an awfully good inkling once you get your first glimpse of the drive itself. Please don’t get us wrong, its not a overly flashy looking drive but it will get your attention. We will get back to the top of the X-25M in a moment but lets first look at the back.

bottom_sm.jpg

The back of the drive is totally and completely devoid of any labels, markings, logos or anything whatsoever. However, on the upside, the case itself is made from solid metal and has a very durable feel to it. Also on the positive side, the Intel engineers did go for bottom and side mounting screw holes in "silver" inserts which not only add a subtle splash of spice to the looks of the drive but also improve add durability and rigidity. We highly doubt you will strip out one of these screw mounts with anything less an butt load of abuse.

top_ang_sm.jpg

Going back to the front of the drive we quickly and readily see that this is where Intel’s designers spent most of their time and effort. The first notable feature is the top of this case is not a flat surface but in fact has a gracefully curving arch embossed into it. This arch is further highlighted and enhanced by the white label which follows this curve. It certainly is different and it does work.


label_sm.jpg

The label itself is filled with all the usual information which you would find on any drive, but one thing really did stand out. In the official specifications (PDF) found on Intel’s own website this unit is rated to run at a heck of a lot less than 1 FULL Amp on the 5v line (5W); yet here it is bold and as clear as day stating its max power consumption is 1 Amp. In most circumstances the X-25M probably uses a heck of a lot less juice than this (closer to the 150mA stated in Intel's spec sheets) but when the label and the "official specifications" differ we always side with the label on the drive.
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,274
Interior Impressions

Interior Impressions


As they say beauty is only skin deep….but ugly goes to the bone. It's time to see if this wee drive is as beautiful on the inside as it is on the outside so let’s pry the top off.

screw_sn.jpg

Doing this is simplicity itself as the top of the casing is held in place with only four ultra small screws. Be warned: removing these screws does nullify your warranty. This may not matter to us as our sample has no warranty to void, but your brand new X-25M does have a 2 year warranty and it would be a shame to void it just to satisfy your curiosity.

pcb_sm.jpg
pcb2_sm.jpg

While the PCB may seem upside down when compared to our previous SSD review’s layout, as the X-25s controller and cache are located on the back side; but once you get past this you quickly realize that Intel didn’t try to totally reinvent the wheel. At one end, you have your SATA power and data connectors, with the SSD controller chip in front and then 2 rows of 5 Intel NAND MLC flash chips. The top side or "other side" of this PCB holds another 2 rows of 5 chips soldered to that side of the board; making it for a grand total of 1 controller chip, 1 RAM chip and 20 NAND flash chips.

controller_sn.jpg

The controller of the X-25M is an Intel System On a Chip pc29AS21AA0 controller. This is the heart and soul of this unit and is the secret to its success. In a nut shell this SOC features advanced architecture which allows it to control 10 parallel NAND flash channels and (with NCQ) can handle up to 32 concurrent operations. The scary thing about this drive is the "M" stands for mainstream and as such is not considered by Intel to be an extreme performance drive.

The easiest way to think about how this controller and NAND flash is set up is to not think about RAID 0 drives (after all, SSDs do not need to write and read in a linear fashion for maximum performance); rather think about how your computer’s memory controller runs. In modern systems you can have dual or triple channel memory all controlled by one controller which can read and write to any of the channels at the same time. Now take that triple channel setup and make it a TEN channel setup, swap out the i7 memory controller for the X-25M's SOC controller, and swap out the DDR3 RAM for MLC NAND Flash RAM. THIS is what Intel has done and to be honest with you, it really is a work of art. The very fact they managed to cram this all into 2.5" laptop form factor just increases the wow factor.

chips_sn.jpg

This 50nm, 8GB (32Gbit) MLC NAND flash model number 29F32G08CAMC1 is also made by Intel. Not much is known about this memory as Intel is very tight lipped about their specifications but they do state it meets the Open NAND Flash Interface 1.0 specifications. This is not that surprising as Intel is a member of the ONFI consortium which are working to develop open standards for NAND flash chips. In any case there are a grand total of 20 on the PCB all of which are grouped into pairs unto the 10 parallel data lanes.

cache_sm.jpg

As you can see in the picture above, the Intel X-25m has onboard cache. However, unlike most drives the cache is solely used for caching commands sent to and from your motherboard's onboard ATA controller. In a nut shell this RAM chip allows the X-25M to keep your system stutter free even when it is overloaded by keeping the ATA controller happy. Most likely instead of extended pauses (i.e. your sytem becoming unresponsive) it will just get slower and slower as the demands increase....just like a regular high performance spindle based drive. The chip shown above is a Samsung K4S281632I-UC60. In this particular instance, this chip was made in the 16 week of 2008. Using the online Samsung model decoder this chip is a 16MB (128Mbit), 4 bank 3.3volt, lead free 6ns (166mhs @ CL 3) DRAM chip.

All in all, the X-25M looks a lot like many Samsung based drives on the market today; albeit with a few Intel tweaks added it. Of course (with the exception of the cache chip) this doesn’t have any Samsung chips to be seen anywhere and is a Intel in house design. This not necessarily a bad thing, as Samsung based drives are having stutter issues left, right and center and we highly doubt Intel would allow that to happen in their first foray into SSDs…though, to be honest it wouldn’t be the first time a big company messed up entering a different market niche. Well, that’s enough talking and analysing this drive...let’s find out whether or not the X-25M can walk the walk like it can talk the talk!
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,274
Testing Methodology

Testing Methodology


Testing a hard drive or SSD for that matter is not as simple as putting together a bunch of files, dragging them onto folder on the drive in Windows and using a stopwatch to time how long the transfer takes. Rather, there are factors such as read / write speed and data burst speed to take into account. There is also the SATA controller on your motherboard and how well it works with SSDs to think about as well. For best results you really need a dedicated hardware RAID controller w/ dedicated RAM for SSDs to shine. Unfortunately, most people do not have the time, inclination or monetary funds to do this. For this reason our testbed will be a more standard motherboard with no mods or high end gear added to it. This is to help replicate what you the end user’s experience will be like.

Even when the hardware issues are taken care of the software itself will have a negative or positive impact on the results. As with the hardware end of things, to obtain the absolute best results you do need to tweak your OS setup; however, just like with the hardware solution most people are not going to do this. For this reason our standard OS setup is used. However, except for the XP load test times we have done our best to eliminate this issue by having the drive tested as a secondary drive. With the main drive being a WD 320 single platter drive.

For these tests we used a combination of the ATTO Disk Benchmark, HDTach, HDTune, Cystal Disk Benchmark, h2benchw, SIS Sandra Removable Storage benchmark, PCMark 05 Hard drive tests and IOMeter for synthetic benchmarks.

For real world benchmarks we timed how long XP and Adobe CS3 (w/ enormous amounts of custom brushes installed) startup took, how long a single 4GB rar file took to copy to and then from the hard drives, then copy to itself. We also used 1gb of small files (from 1kb to 20MB) with a total 2108 files in 49 subfolders.

For the temperature testing, readings are taken directly from the hottest part of the drive case using a Digital Infrared Thermometer. The infrared thermometer used has a 9 to 1 ratio, meaning that at 9cm it takes it reading from a 1 square cm. To obtain the numbers used in this review the thermometer was held approximately 3cm away from the drive and only the hottest number obtained was used.

Please note to reduce variables the same XP OS image was used for all the hard drives.

For all testing a Gigabyte PA35-DS4 motherboard was used. The ICH9 controller on said motherboard was used.

All tests were run 4 times and average results are represented.

You will also notice that in a number of benhcmarks the G.Skill SSD is lacking results of is not represented at all. This is due to the fact that we no longer have the drive in our posession and as such could not rerun the tests after we changed our benchmark suite.

Processor: Q6600 @ 2.4 GHZ
Motherboard: Gigabyte p35 DS4
Memory: 4GB G.Skill PC2-6400
Graphics card: Asus 8800GT TOP
Hard Drive: 1x WD 320
Power Supply: Seasonic S12 600W Performance Testing
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,274
Crystal DiskMark

Crystal DiskMark


Crystal DiskMark is designed to quickly test the performance of your hard drives. Currently, the program allows you to measure sequential and random read/write speeds; and allows you to set the number of test iterations to run. We left the number of tests at 5. When all 5 tests for a given section are run, Crystal DiskMark then averages out all 5 numbers to give a result for that section.

READ

CDM_read.jpg

This beast really is capable of some monster read performance! While the sequential and 512 numbers are down right amazing (nearly 2 or 3 times greater than the next best IS amazing!!) but what was really impressive was the small 4k read test. That number is down right amazing. Honestly, when you see numbers which are for all intents and purposes TWENTY times better than a VelociRaptor (well actually more like 19.5 times but who is counting?), it becomes hard to find words to accurately describe performance.


Write
CDM_write.jpg


As we saw with the HD Tune Write test, the X-25 is only average when it comes to sequential and / or 512k chunks writes; BUT when the file size shrinks down to 4k chunks…then it’s a Déjà vu all over again, with (better than) 20x better performance numbers! As with the read section the X-25M’s controller and MLC NAND chips shine with small files, and in the real world this is what is really important as your OS darn near constantly reads and writes little data chunks to and from your OS drive.
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,274
Read and Write Performance

Read Bandwidth


For this benchmark, HDTach was used. It shows the potential read speed which you are likely to experience with these drives. The long test was run to give a slightly more accurate picture.

We don’t put much stock in Burst speed readings and this goes double for SSD based drives. The main reason we include it is to show what under perfect conditions a given drive is capable of even though the more important result is that of the Average Speed. This number will tell you what to expect from a given drive in normal, day to day operations. The higher the average, the faster your entire system will seem.


read.jpg

I honestly didn’t believe these numbers were correct the first time I ran HDTach; however by the third try I was becoming a believer and when the 4th came up the same….well how else can you describe read speeds which are litteraly more than DOUBLE that of a VelociRaptor or WD 640GB? Term like "insane" and some others we can’t print come to mind, but I think eventhe guy upstairs won’t mind me taking his name in vein when s/he sees these numbers!


Write Performance


For this benchmark HD Tune Pro was used. To run the write benchmark on a drive, you must first remove all partitions from that drive and then and only then will it allow you to run this test. Unlike some other benchmarking utilities, HD Tune Pro writes across the full area of the drive, thus it easily shows any weakness a drive may have.

While most OS drives spend most of their times reading and not writing, the write speed of the drive does have a big impact on the stutter issue and how fast the drive feels.


write.jpg

The X-25M is "only" capable of an average of 72 MB/s writes but what is more important was its minimum was a rock solid 64.5 MB/s. This is very, very good and it is highly unlikely that this drive will suffer and stuttering due to slow writes. It may suffer from them IF the controller and cache buffer get overwhelmed BUT it won’t be because those 20 MLC NAND chips weren’t performing up to spec.
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,274
Random Access Time / SiS Sandra

Random Access Time


To obtain the absolute, most accurate Random access time, h2benchw was used for this benchmark. This benchmark tests how quickly different areas of the drive’s memory can be accessed. A low number means that the drive space can be accessed quickly while a high number means that more time is taken trying to access different parts of the drive. To run this program, one must use a DOS prompt and tell it what sections of the test to run. While one could use "h2benchw 1 -english -s -tt "harddisk test" -w test" for example and just run the seek tests, we took the more complete approach and ran the full gamut of tests and then extracted the necessary information from the text file. This is the command line argument we used "h2benchw 1 -a -! -tt "harddisk drivetest" -w drivetest". This tells the program to write all results in english, save them in drivetest txt file, do write and read tests and do it all on drive 1 (or the second drive found, with 0 being the OS drive).

random.jpg


When you get into numbers this small they really start to become almost abstract in their size. What we mean is 0.09 is basically a smaller slice of time than we can precieve, heck it takes about .33 of second to blink your eye…and the X-25 is 3.6 times faster than that! If that is not impressive enough for you consider this: Bob Munden who is considered the "fastest gun in the world" can draw and fire a SAA colt in .27 seconds…and this X-25M would eat him for lunch (and leave him dead in the ground in a draw). So respect the X-25M ‘cause it deserves it.


SIS Sandra


This test was run with the removable storage benchmark in Sandra XII Standard. All of the scores are calculated in operations per second and have been averaged out from the scores of 4 test runs.


sis.jpg


It is becoming an almost monotonous reoccurring theme; the Intel X-25 simply stomps all over the competition in the SIS benchmark department. Numbers these large were simply unimaginable a short while ago.
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,274
ATTO Disk Benchmark

ATTO Disk Benchmark


The ATTO disk benchmark tests the drive's read and write speeds using gradually larger size files. For these tests, the ATTO program was set to run from its smallest to largest value (.5KB to 8192KB) and the total length was set to 256MB. The test program then spits out an extrapolated performance figure in megabytes per second.

Read

ATTO_READ.jpg


The X-25M starts out in first place and as the test continues simply widens its lead over the rest of the competition. It really is a monster when it comes to read speeds and we highly doubt we are going to see ANY single drive beat it in the near future. That right there is a bold statement and we may end up eating our words but we truly doubt it will be happening anytime soon…but we look forward to seeing what does come along to eat it for lunch.


Write
ATTO_WRITE.jpg

Write speed is obviously the weak link in the X-25M’s chain. It may start out in first but as the size of the files increase the speed of this drive slows down in its upward climb which eventually lands it in dead last place. What is also obvious is Intel has a specific goal in mind when they built and tweaked this controller and that was to optimize small file writes. To us this makes perfect sense as in the real world, small files are the norm and on large files no one really cares if it takes an extra couple seconds to finish as you are going to have to wait either way.

We hope other companies (like JMicron) take note of this and opt to follow suit as synthetic numbers are all well and fine but it’s the real world performance which counts the most.
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,274
IOMETER / Stutter

IOMETER


IOMeter is heavily weighted towards the server end of things, and since we here at HWC are more End User centric we will be setting and judging the results of IOMeter a little bit differently than most.

To test each drive we ran 5 test runs per drive (1,4,16,64,128 que depth) each test having 8 parts, each part lasting 10 min w/ an additional 20 second ramp up. The 8 subparts were set to run 100% random, 80% read 20% write; testing 512b, 1k, 2k,4k,8k,16k,32k,64k size chunks of data. When each test is finished IOMeter spits out a report, in that reports each of the 8 subtests are given a score in I/Os per second. We then take these 8 numbers add them together and divide by 8. This gives us an average score for that particular que depth that is heavily weighted for single user environments.


IOMETER.jpg


The X-25M is a monster when it comes to IOMeter results and it actually is so much better that it crams all the others at the bottom of the graph. This is far from an easy test, yet the X-25M makes it look easy by posting monster numbers.


IOMeter Stutter Test


In our usual IOMeter test we are trying to replicate real world use where reads severely outnumber writes. However, to get a good handle on how well a Solid State Drive will handle a worse case scenario (and thus how likely the dreaded stutter issue will happen) we have also run an additional test. This test is made of 1 section at que depth of 1. In this test we ran 100% random and included 100% writes of 4k size chunks of information. In the .csv file we then found the Maximum Write Response Time. This in ms is worst example of how long a given operation took to complete. We consider anything higher than 333ms (one third of a second) to be a good indicator that stuttering may happen, with the higher the number the longer the duration of the stutter will most likely be.

Stutter.jpg

When it comes to predicting stutters, there is good news and mediocre news when it comes to the X-25M. The good news is the average write time is amazingly good; the mediocre is .322 of a second is only mediocre at best. It is still good enough to minimize stutters from happening (and this drive does do a heck of a lot more in a second than most drives do in 5) but more importantly we have a sneaking suspicion that this number is going to stand as the gold standard for MLC SSDs for the next while as well. Only time and more SSDs will prove to us one way or the other but to us these numbers are good enough to earn a pass from us.
 
Last edited by a moderator:
Status
Not open for further replies.
Top