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!

G.Skill Falcon 128GB SSD Review

Status
Not open for further replies.

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,270
gskill_logo-1.png


G.Skill Falcon 128GB SSD Review




Manufacture Page: Falcon
Part Number: FM-25S2S-128GBF1
TechWiki Info: Falcon
Price: Click Here to Check Prices
Warranty: 2 years


It appears that after decades of seeming stagnation, the storage sector is finally starting to shift gears. For the first time in computer history the lowly storage device is starting to mature at a rate which destroys Moore's Law. This change in pace is not happening in the regular spindle based Hard Disk Drives (though one could argue that the density increases has been just as fast as CPU speed increases); rather, it is happening in the Solid State Device sector. The SSDs of just a few months ago are already becoming mundane in their speeds and capacities and there is already a whole new generation set to supersede anything we have seen before. Will this increase in speed AND size continue on an annual or semi-annual basis? If it does, it will most likely be more chaotic than the ATI / Nvidia “war” or the AMD / Intel one. While those industries have it down to a “tick-tock” cycle it appears the SSD marketplace is still finding its legs and is actually more cut throat than ever.

Just a little while ago an unknown company burst onto the scene. The Barefoot controller by Indilinx really has stuck a fork in the dying JM controller lines and introduced a 3rd serious player in the SSD chipset market. However, unlike Intel or Samsung controllers it appears that Indilinx is extremely price competitive for the performance it offers. This assumption is based on the fact that so soon after the initial launch, G.Skill has come along with their Falcon line to undercut the price strategy that OCZ had in place. If the G.Skill drive proves to be just as good as any out there, many others may soon jump on the Barefoot controller bandwagon. This would be great for consumers as competition breeds price-cuts!

Indeed, this article will allow us to see if the G.Skill Falcon 128GB SSD can really put the rubber to the road so to speak. Since the Barefoot has been out there for awhile and a lot of R&D and tweaking has gone on, the Falcon may not suffer any growing pains like a certain other drive did….and make all those early adopters nothing more than unpaid guinea pigs for G.Skill (and any other manufacturer willing to jump on the Barefoot bandwagon). Let’s find out if our initial assumptions hold water and just how good this bad boy is.

GSKILL_FALCON_FALCON_1.jpg

 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,270
Specifications

Specifications


<img src="http://images.hardwarecanucks.com/image/akg/Storage/Falcon128/GSKILL_FALCON_spec.jpg" border="0" alt="" />
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,270
Packaging and Accessories

Packaging and Accessories


GSKILL_FALCON_box_front_sm.jpg
GSKILL_FALCON_box_bac_sm.jpg

When we first laid eyes on the package the Falcon was ensconced in we were a little disappointed. In all honesty, the Titan colour scheme is a much better fit for a drive which promises to be as powerful as this SSD should be. However, once we got over our disappointment and took a second longer look at this box’s colour scheme it did start to grow on us. Where the Titan is all raw machismo and reminded us a lot of an Outlaw biker, this box screams elegance and agility, much like a soaring eagle….or Falcon!

GSKILL_FALCON_box_open_sm.jpg

When we did finally open the box (after reading all the minutia and data included on the back of the box first) we were greeted to the exact same internal protection scheme as the Titan. This large high density foam protection (which still reminds us a lot of a book which has been hollowed out) works very well indeed.

GSKILL_FALCON_access_sm.jpg

When we removed the Falcon from its interior protector, we noticed a very interesting accessory was included: a two pin jumper. As most enthusiast are well aware of, the OCZ Vertex is firmware updateable and it appears that this feature is going to be carried over to any and all Indilinx Big Foot Controller SSDs. This is great news as it means that if OCZ has a breakthrough in firmware tweaks….it will eventually trickle down to G. Skill’s Falcon (and vice versa!). One strange thing was our Falcon did not come with any instruction pamphlet and while this is not a big loss it was conspicuous by its absence.
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,270
First Impressions

First Impressions


GSKILL_FALCON_bottom_ang_sm.jpg
GSKILL_FALCON_bottom_side_sm.jpg

As with the Titan SSD, the Falcon is wrapped in an all black metal body. Also like the Titan the “top” of this drive is denude of any goo-gaws, glow in the dark decals, or the most unholy of unholies the holographic sticker. The only thing on the top of this drive is a large sticker which is for all intents and purposes a clone of the packaging box colour and logo scheme. In a nut shell, it tells you what this drive is, who makes it and the size and form factor of the drive; no extraneous boasts or claims, just pure information in a low key sticker. In many ways the white and black colour works better than the black and red of the titan as the Falcon’s does stand out more. In either case, we like and respect its simplicity and quickly ignored it once we absorbed its information….just like any properly designed sticker should be!

GSKILL_FALCON_top_ang_sm.jpg
GSKILL_FALCON_top_sm.jpg

The back label is your fairly typical, standard fare with its white background and black lettering. As with all SSDs, including the Falcon, this label is crammed full of useful information. The biggest of these is the 64MB onboard cache listed. This right here is one of the things we have been looking for in SSDs and so far only Intel, Samsung and OCZ have delivered. The inclusion of all that beautiful cache should darn near ensure a stutter free experience, assuming previous experiences are anything to go by.

GSKILL_FALCON_jumper_sm.jpg

The only other noticeable feature of the Falcon is located adjacent the SATA data and power ports. As we mentioned in the accessories section this particular drive has completely upgradeable firmware.

To perform a firmware upgrade you use the included jumper to jump these two small pins allowing it to go into its updateable mode. Its a fast and simple yet darn near idiot proof way of doing this…and as a bonus if you don’t know what those jumpers are, its very unlikely you will ever mess with it. The only downside to setting a jumper to flash the FW is if an end-user “has enough information to be dangerous” and is used to PATA drives. There is the possibility they will think this is the jumper to turn their SATA drive into a “slave” drive. If you are giving this drive as a gift, or setting up a system for a less knowledgeable user…either don’t tell them about this all together or explain that “bad things happen” if they do try it without you. In all honesty we are glad to see the jumper abilities on the Falcon and think this setup is the best compromise possible.

All in all we walked away with a very, very, VERY good first impression of the Falcon. It looks good, is well protected and G.Skill has obviously put a lot of time and effort into the packaging of this monster machine. However, one question does push its way to the fore: Will the Falcon soar and live up to our lofty expectations or will it on closer examination turn into a grounded turkey? Before we get to performance however, let’s take a close look at it and then crack it open like a lobster to get to the good bits.
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,270
Interior Impressions

Interior Impressions


Before we continue: Please remember that opening any SSD will effectively void your warranty.

GSKILL_FALCON_void_sm.jpg

To open the Falcon up you have to remove four screws, which is par for the course; what is unusual is we had to remove not one but TWO “warranty void” warning stickers to get at the screws. We honestly don’t think anyone will ever mistakenly take apart a Falcon with this kind of obvious warnings. While this was interesting in an annoying way, it did cross over into the surreal when we spotted a THIRD warning label…inside the case.

On first glance at the board, one could easily be excused for thinking this was another Titan clone (I.E. dual JM controllers). Its only when you take a close look to do you realize that one of the chips is the Barefoot controller and the other is the Ram chip.

GSKILL_FALCON_top_off_sm.jpg
GSKILL_FALCON_bottom_board_sm.jpg

Other than this Ram and Controller tweak this board looks a lot like any other board out there. At one end you have the data and power ports (and the ingenious jumper pins), then the RAM and Barefoot controller in from it followed by your typical double row of flash chips, with another double row of 4 on the other side. In grand total you have 16
flash chips, one DRAM chip and one Indilinx Barefoot controller chip.

GSKILL_FALCON_control_sm.jpg

The I/O controller chip is (as expected) none other than the Indilinx “Barefoot” IDX110 controller. The Indilinx IDX110 is an ARM based controller with native SATA 3.0Gb/s, supports capacities of up to 512GB (vs. 256 for the FC version) and is rated for speeds of 230MB/Sec read and 190MB/s writes. Also and just as importantly, unlike the JM line of controllers this bad boy also supports up to 64MB of RAM for stutter free operation. On paper this is certainly a potent brew and we are itching to see how it fares in the both the synthetic and real world tests!

This controller really is the secret sauce that makes the Falcon so darn unique; and in an interesting twist this is the Indilinx IDX110M00-LC (with emphasis on the LC) and not the older -FC found in previous competitors Barefoot SSD. To put this another way this is the brand new version of the IDX110 “barefoot” controller which has up until now only been seen in the uber expensive Vertex EX line. We have a feeling that the FCs have been phased out and all new Barefoot based SSD will also have this revision. In any case it was a nice little surprise and bonus.

GSKILL_FALCON_ram_sm.jpg

The Ram which helps the Barefoot go stutter free is made by Elpida. To be specific this single 64MB SDRAM chip is S51321CBH-6DTT-F, though the actual Elpida part number is the EDS51321CBH-M-F (but we assume this was too much to fit on this little chip). This ram chip is rated to run at 166MHZ at CL3 and is rated for an operating temperature range of 20°C to 85°C.

GSKILL_FALCON_flash_sm.jpg
GSKILL_FALCON_flash2_sm.jpg

The MLC NAND chips used in the Falcon are Samsung K9HCG08U1M-PCB0 units. Once again, using the online Samsung model decoder we can see these chips are 48 pin 3bit MLC Quad Die Package, 1st gen lead free (ROHS compliant), 2.7V ~ 3.6V, 25 nanosecond NAND chips which operate with Dual nCE (Dual Chip Enable control) & Dual R/nB (Dual Ready/Busy Output). This model is rated at a density of 64Gbits or 8GB per chip and for temperatures for 0° to 70°C. Above the model number (and as stated earlier) we can see these were made in the 10th week of 2009 and below it we can see the batch number “YC84A1XX” (or at least what we assume is the batch number but is describe by Samsung vaguely as “Customer List Reference” only).
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,270
TRIM / Wiper.exe

TRIM and Firmware Update


As this review was in the process of going live, a newer firmware was released. It is mainly a bug fix release so upgrading the firmware is not really necessary as all Falcons come with Firmware 1.10 (1370) and are TRIM command capable right out of the box. We of course reran all the benchmarks before and after and there was no noticeable difference in performance. In the end it is up to you if you think it is worth doing; however with that being said, we do strongly recommend you stop by every once in awhile and see if the official Falcon Firmware thread on their forum contains any new information or updates. Here is a link to the official 1571 thread: [ONLY for FALCON SSD]Firmware 1571 released - GSKILL TECH FORUM before running this firmware we strongly urge you to download the manual and read it.

wiper02523_sm.jpg

As we said, updating the firmware is not as crucial as it is with other Barefoot-based SSDs as all Faclon's support the TRIM command right out of the box. While support for the Trim command is all well and fine, unless you have Win 7 you would be out of luck….or at least you would be if not for a handy little program called “wiper.exe”. The official thread for G.Skill’s version is here: [only for Falcon Series]wiper.exe - TRIM COMMAND for Falcon Series SSD - GSKILL TECH FORUM. We recommend you also stop by from time to time as the wiper program is still beta.

The most important thing to realize is this is beta firmware and while the chances of data loss are slim you must be cognizant of this possibility. G. Skill’s wiper program bears an awfully suspicious resemblance to OCZ’s original version of their formatting program and as such if you are using a 64bit OS you may wish to wait a bit longer before using it….or backing up all your data just in case. On an interesting side note, all OCZ’s wiper versions appear to work just fine on the Falcon. OCZ states the original version is about 50/50 chances of data loss on 64bit programs but only 10% on 32bit programs. The never 0422 has its detractors as some say its not as good at its job as the original but they say its darn near 100% on 32bit OS’ and 90% safe on 64bit. Food for thought on which program you should use to say the least.

wiper01623_sm.jpg

To run the wiper program, the best thing is to have the Falcon be the OS drive and drive 0. To properly run, it needs the Falcon to be Drive 0 but it seemed more hit or miss when it was a data (aka D drive). Since using a Falcon (or any SSD) as a data drive is a little unlikely, this caveat is a minor one to say the least. The only other restriction is it will not work on Falcon’s which have been RAID’ed.

To run it all you need to do is double click the wiper.exe icon. It should be located in the root folder. When you double click it, a DOS window opens up and it runs all by itself. If you have made multiple partitions, you need not worry about running it for each partition as the program is smart enough to look for ALL Falcon drives and run itself on all “drives”. While we only have one Falcon to play with, we assume that as long as you have not RAID’ed multiple Falcons it will also wipe any and all of them in one fell swoop, but this is only supposition. If you have multiple non Falcon drives you need not worry as it will simply tell you it cannot run on XYZ drive and then continue on looking for the Falcon, find it and then clean it and then finish.

We can tell you that this program does work as it restored speed to the drive after we had finished torturing it with IOMeter. With Crystal DiskMark the scores were down right abysmal after IOMeter, but after running the wiper program they were right back up where they were before….like nothing had happened. As with all Indilinx SSDs, you should run any and all benchmarking BEFORE installing the OS. This way you can have your fun, wipe it clean and then use it with no adverse effects.
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,270
Testing Methodology

Testing Methodology


Testing a hard drive 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, and IOMeter for synthetic benchmarks.

For real world benchmarks we timed how long XP startup took, Adobe CS3 (w/ enormous amounts of custom brushes installed) 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 heatsink 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.

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


SSD FIRMWARE (unless otherwise noted):

G. Skill Titan: 0955
G.Skill Falcon: 1571
OCZ Apex: 955
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,270
Read Bandwidth / Write Performance

Read Bandwidth


For this benchmark, HDTach was used. It shows the potential read speed which you are likely to experience with these hard 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 hard drive. The main reason we include it is to show what under perfect conditions a given drive is capable of; but the more important number is the Average Speed number. 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.


GSKILL_FALCON_read.jpg


With an average speed this high, you KNOW the Falcon is a fast drive! An average speed of darn near 230MB/s from a single “hard drive” would have been considered impossible just a few years back and yet here we are!

The lower burst speed is par for the course with Indilinx drives but it may have more to do with that particular controller not getting along with the way HDTach records burst speeds than an out and out issue with the Barefoot controller per say. In either case, burst speeds are not that important and even if they were this drive's burst speed is still awfully good and only looks “bad” when compared with its monstrously huge average.


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 utlities the 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.


GSKILL_FALCON_write.jpg


194MB/S average write speed is even more phenomenal than the read speed numbers this monster posted! Even better than the average was the minimum speed which is almost more than double that of the mighty WD VelociRaptor and is more than double that of the Intel X-25M. If you do not find that impressive…turn in your PC enthusiast badge as you must have cheated on the super secret entrance exam.
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,270
Crystal DiskMark

Crystal DiskMark


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

READ

<img src="http://images.hardwarecanucks.com/image/akg/Storage/Falcon128/GSKILL_FALCON_CDM_read.jpg" border="0" alt="" />

When it comes to ultra small file reads, this drive easily beats the Intel X-25M and in the other two tests comes awfully darn close to tying it. It is obvious that the Barefoot controller w/ firmware 1370 is tweaked for small file transfer speed. This is exactly what you want to see in any OS drive as this is what most of the background reads are going to consist of. Bravo Indilinx for being only the third company to catch the self same clue stick which has been beating JMircon over the head now for a long….long while!

Write

<img src="http://images.hardwarecanucks.com/image/akg/Storage/Falcon128/GSKILL_FALCON_CDM_write.jpg" border="0" alt="" />

The write speeds are also impressive but they do highlight the different philosophies between Intel and Indilinx. While 11+ MB/s on ultra small files is impressive it appears the Barefoot controller is designed to handle larger files better; where as the Intel is all about the small file speed (even to the detriment of larger chunk write performance). This may just be a result of a relatively early firmware which has not been dialed in, as the larger chunks of data performance are down right impressive. We would not be overly surprised if future firmware toned down the high end but increased the low end performance. As with everything in life, it's all about balance and it certainly will be interesting to see how this particular controller handles real world situations with its curious mix of read and write speeds.
 
Last edited by a moderator:

AkG

Well-known member
Joined
Oct 24, 2007
Messages
5,270
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 gamout 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).

<img src="http://images.hardwarecanucks.com/image/akg/Storage/Falcon128/GSKILL_FALCON_random.jpg" border="0" alt="" />

For all intents and purposes this drive is just as fast as the Intel X-25. Honestly, when you get down to differences of 0.01ms you really are cutting hairs and looking awfully hard for nits to pick. We consider both the Indilinx based Falcon and the Intel X-25M to be in a league all their own, and masters of a field which is known for ultra low latencies!


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.

<img src="http://images.hardwarecanucks.com/image/akg/Storage/Falcon128/GSKILL_FALCON_sis.jpg" border="0" alt="" />

For whatever reason, the fuzzy logic of SISoft does not like this drive as much as some others. Is that a big deal? No, as this is easily the most schizophrenic test ever invented. Run this “test” 10 times and you will get a different answer every time. Needless to say the results are not bad, but in reality it’s like worrying about who placed 3rd in the Special Olympics. Sure it’s important to the people involved but for everyone else….who cares!
 
Last edited by a moderator:
Status
Not open for further replies.

Latest posts

Top