SSD general ?, and SSD in a 2008 2.8 8-core?
- Thread starter snouter
- Start date
- Sort by reaction score
iamcheerful
macrumors 6502
My experiences
SSD certainly is a recommended upgrade (only if you care to fork out that kind of budget $).
I haven't touched one or tried one. However, I've seen demo online (youtube, the comparison test AppleInsider did then with the MacBook Air).
The first hand experience I have similar to that of SSD is RAM Disk. I'm sure many veteran Apple Users know about this. I for one was kinda sad that I wasn't able to boot from a RAM Disk with my (now resting in peace) TiPB G4 400MHz. We won't talk about loading Flash sites as that is quite CPU intensive too. However for the usual typical surfing in the early 2000s (i.e. 2001, 2002 etc), my PB 3400c/200MHz could hold her own (in fact, was faster even) when surfing from RAM Disk. The complete OS was installed in the RAM Disk and everything just flew. Moving back and forth pages was a breeze. The TiPB couldn't do that w/o employing a trick (i.e. moving certain files etc to a RAM Disk which wasn't bootable).
Booting from a RAM Disk was FAST. It was silent too! No fan noise at all. No hdd spin ups or downs. Typing or authoring a book was extremely comforting.
One bad thing though ... the PB3400 was a heavy machine! lol I would have to type with a table. I could use it as a laptop but after some time, it would be advisable to move to a proper sitting position.
All said, I believe SSD is pretty similar to a RAM Disk.
p.s. Battery life was amazing too. I could get up to 4.5 hrs (this is a conservative estimate as it has been years since I last lay my hands on that machine) of battery life on the PB3400c when I booted up with RAM Disk.
SSD certainly is a recommended upgrade (only if you care to fork out that kind of budget $).
I haven't touched one or tried one. However, I've seen demo online (youtube, the comparison test AppleInsider did then with the MacBook Air).
The first hand experience I have similar to that of SSD is RAM Disk. I'm sure many veteran Apple Users know about this. I for one was kinda sad that I wasn't able to boot from a RAM Disk with my (now resting in peace) TiPB G4 400MHz. We won't talk about loading Flash sites as that is quite CPU intensive too. However for the usual typical surfing in the early 2000s (i.e. 2001, 2002 etc), my PB 3400c/200MHz could hold her own (in fact, was faster even) when surfing from RAM Disk. The complete OS was installed in the RAM Disk and everything just flew. Moving back and forth pages was a breeze. The TiPB couldn't do that w/o employing a trick (i.e. moving certain files etc to a RAM Disk which wasn't bootable).
Booting from a RAM Disk was FAST. It was silent too! No fan noise at all. No hdd spin ups or downs. Typing or authoring a book was extremely comforting.
One bad thing though ... the PB3400 was a heavy machine! lol I would have to type with a table. I could use it as a laptop but after some time, it would be advisable to move to a proper sitting position.
All said, I believe SSD is pretty similar to a RAM Disk.
p.s. Battery life was amazing too. I could get up to 4.5 hrs (this is a conservative estimate as it has been years since I last lay my hands on that machine) of battery life on the PB3400c when I booted up with RAM Disk.
300D
macrumors 65816
This is my old G4 in xbench a few months ago using a low-end SSD.
System Info
Xbench Version 1.3
System Version 10.4.11 (8S165)
Physical RAM 2048 MB
Model PowerMac3,1
Processor PowerPC G4 @ 600 MHz
Version 7410 (Nitro) v1.3
L1 Cache 32K (instruction), 32K (data)
L2 Cache 1024K @ 300 MHz
Bus Frequency 100 MHz
Video Card ATY,Rage128Pro
Drive Type Maxtor 90845D4 (8GB)
Disk Test 17.42
Sequential 17.27
Uncached Write 9.70 5.95 MB/sec [4K blocks]
Uncached Write 18.89 10.69 MB/sec [256K blocks]
Uncached Read 35.31 10.33 MB/sec [4K blocks]
Uncached Read 21.15 10.63 MB/sec [256K blocks]
Random 17.58
Uncached Write 7.03 0.74 MB/sec [4K blocks]
Uncached Write 24.14 7.73 MB/sec [256K blocks]
Uncached Read 51.94 0.37 MB/sec [4K blocks]
Uncached Read 40.69 7.55 MB/sec [256K blocks]
Drive Type PATRIOT MEMORY 32GB SSD
Disk Test 62.14
Sequential 68.31
Uncached Write 93.00 57.10 MB/sec [4K blocks]
Uncached Write 104.47 59.11 MB/sec [256K blocks]
Uncached Read 33.19 9.71 MB/sec [4K blocks]
Uncached Read 123.36 62.00 MB/sec [256K blocks]
Random 56.99
Uncached Write 18.06 1.91 MB/sec [4K blocks]
Uncached Write 92.66 29.66 MB/sec [256K blocks]
Uncached Read 1155.01 8.18 MB/sec [4K blocks]
Uncached Read 318.10 59.03 MB/sec [256K blocks]
Thats a 3.5x performance jump. It made a night and day difference in performance. The computer that was once slow and laggy became quite snappy and usable.
Granted, its an old G4 and not a Mac Pro but, again, its the latency that makes the most noticeable difference and not the throughput.
7ms is 0.007 seconds but an SSD is 0.00001 seconds, 1/700th the time. And thats consistent across the entire memory range, it doesn't vary like a hard drive.
System Info
Xbench Version 1.3
System Version 10.4.11 (8S165)
Physical RAM 2048 MB
Model PowerMac3,1
Processor PowerPC G4 @ 600 MHz
Version 7410 (Nitro) v1.3
L1 Cache 32K (instruction), 32K (data)
L2 Cache 1024K @ 300 MHz
Bus Frequency 100 MHz
Video Card ATY,Rage128Pro
Drive Type Maxtor 90845D4 (8GB)
Disk Test 17.42
Sequential 17.27
Uncached Write 9.70 5.95 MB/sec [4K blocks]
Uncached Write 18.89 10.69 MB/sec [256K blocks]
Uncached Read 35.31 10.33 MB/sec [4K blocks]
Uncached Read 21.15 10.63 MB/sec [256K blocks]
Random 17.58
Uncached Write 7.03 0.74 MB/sec [4K blocks]
Uncached Write 24.14 7.73 MB/sec [256K blocks]
Uncached Read 51.94 0.37 MB/sec [4K blocks]
Uncached Read 40.69 7.55 MB/sec [256K blocks]
Drive Type PATRIOT MEMORY 32GB SSD
Disk Test 62.14
Sequential 68.31
Uncached Write 93.00 57.10 MB/sec [4K blocks]
Uncached Write 104.47 59.11 MB/sec [256K blocks]
Uncached Read 33.19 9.71 MB/sec [4K blocks]
Uncached Read 123.36 62.00 MB/sec [256K blocks]
Random 56.99
Uncached Write 18.06 1.91 MB/sec [4K blocks]
Uncached Write 92.66 29.66 MB/sec [256K blocks]
Uncached Read 1155.01 8.18 MB/sec [4K blocks]
Uncached Read 318.10 59.03 MB/sec [256K blocks]
Thats a 3.5x performance jump. It made a night and day difference in performance. The computer that was once slow and laggy became quite snappy and usable.
Granted, its an old G4 and not a Mac Pro but, again, its the latency that makes the most noticeable difference and not the throughput.
7ms is 0.007 seconds but an SSD is 0.00001 seconds, 1/700th the time. And thats consistent across the entire memory range, it doesn't vary like a hard drive.
Tesselator
macrumors 601
And my ram is:
And my 3-Drive RAID0 is:
So are we on the same page yet?
Code:
_________________________________________________
Benchmark: CPU & RAM
Version: 1.000
Date/Time: 16 June 2009, 07:39:05 PM
Operating System: MacOS 10.57
Total Time: 18,619 millisecs (0:18)
Average Speed: 849.5M bytes/second
Time Speed
Memory Copy: 971 1.1G/sec
Memory Fill: 2,633 1.7G/sec
Memory Clear: 2,638 1.7G/sec
Memory Equal: 1,115 1004.4M/sec
CRC: 3,208 199.5M/sec
Quick Hash: 1,145 558.9M/sec
Secure Hash: 1,541 207.6M/secAnd my 3-Drive RAID0 is:
Code:
+--------------------------------------------------------------------------------------+
| Blocksize | IOPs | Throughput | User CPU | Avg Latency | Max Latency |
+---------------------+----------+--------------+----------+-------------+-------------+
4 K Seq Read 376616.0 1471.156 MB/s 0.734 s 0.019 ms 0.599 ms
Random Read 276111.3 1078.560 MB/s 1.352 s 0.024 ms 1.036 ms
Seq Write 80493.0 314.426 MB/s 0.702 s 0.024 ms 0.729 ms
Random Write 9261.5 36.178 MB/s 0.755 s 0.409 ms 940.026 ms
Create scratch file 43226.3 168.853 MB/s 0.920 s 0.112 ms 0.742 ms
8 K Seq Read 293392.9 2292.132 MB/s 0.321 s 0.025 ms 0.230 ms
Random Read 244531.5 1910.402 MB/s 0.526 s 0.030 ms 0.339 ms
Seq Write 43918.3 343.112 MB/s 0.336 s 0.033 ms 1.219 ms
Random Write 4870.9 38.054 MB/s 0.372 s 1.044 ms 689.109 ms
Create scratch file 29008.2 226.626 MB/s 0.383 s 0.129 ms 0.945 ms
16 K Seq Read 187209.8 2925.153 MB/s 0.193 s 0.040 ms 0.628 ms
Random Read 184884.3 2888.817 MB/s 0.284 s 0.039 ms 0.921 ms
Seq Write 4183.4 65.365 MB/s 0.190 s 1.531 ms 90.564 ms
Random Write 1397.7 21.840 MB/s 0.270 s 4.643 ms 105.407 ms
Create scratch file 7043.7 110.058 MB/s 0.086 s 0.055 ms 0.375 ms
32 K Seq Read 106724.5 3335.140 MB/s 0.102 s 0.067 ms 1.522 ms
Random Read 108085.3 3377.665 MB/s 0.153 s 0.065 ms 0.802 ms
Seq Write 2860.2 89.381 MB/s 0.102 s 2.420 ms 104.268 ms
Random Write 1356.1 42.378 MB/s 0.142 s 4.595 ms 72.960 ms
Create scratch file 9780.7 305.646 MB/s 0.112 s 0.189 ms 0.969 ms
64 K Seq Read 55971.0 3498.185 MB/s 0.055 s 0.127 ms 0.987 ms
Random Read 59866.2 3741.640 MB/s 0.082 s 0.119 ms 0.955 ms
Seq Write 3806.2 237.886 MB/s 0.050 s 1.603 ms 21.540 ms
Random Write 1143.1 71.444 MB/s 0.072 s 4.974 ms 69.624 ms
Create scratch file 4305.2 269.073 MB/s 0.069 s 0.382 ms 2.115 ms
128 K Seq Read 28523.0 3565.380 MB/s 0.030 s 0.251 ms 1.115 ms
Random Read 30432.7 3804.089 MB/s 0.044 s 0.235 ms 1.058 ms
Seq Write 2024.2 253.020 MB/s 0.029 s 2.881 ms 30.079 ms
Random Write 1032.3 129.037 MB/s 0.040 s 5.530 ms 45.979 ms
Create scratch file 2678.0 334.755 MB/s 0.045 s 0.699 ms 6.232 ms
256 K Seq Read 14057.3 3514.321 MB/s 0.023 s 0.519 ms 1.566 ms
Random Read 15110.7 3777.683 MB/s 0.026 s 0.468 ms 1.286 ms
Seq Write 941.8 235.458 MB/s 0.016 s 6.210 ms 46.990 ms
Random Write 656.0 164.011 MB/s 0.022 s 8.436 ms 47.728 ms
Create scratch file 1310.0 327.489 MB/s 0.028 s 1.242 ms 16.358 ms
+--------------------------------------------------------------------------------------+
Benchmark Parameters Summary
============================
Device : / (/dev/disk4), Macintosh RB2@0
Block Size Range
Start: 4 K
End: 256 K
File I/O Size: 10 MB
Number of threads: 8
File System Cache: Off
Read Ahead: Off
IO type: File IO
Tests Run:
Sequential Read
Sequential Write
Random Read
Random WriteSo are we on the same page yet?
Tesselator
macrumors 601
Well it looks like you could have had a 10x performance jump if you had spent the same money on a RAID plus you would have 4.5 terabytes plus you would have 1E15 UBE media instead of 1E4/1E5 for SSD.
Code:
Results 176.78
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 12288 MB
Model MacPro1,1
[U]Drive Type Macintosh RB2 3-Drive RAID0 [B]4.5 Terabytes![/B] [/U]
Disk Test [B]176.78[/B]
Sequential 175.77
Uncached Write 443.19 272.11 MB/sec [4K blocks]
Uncached Write 479.91 271.53 MB/sec [256K blocks]
Uncached Read 60.62 17.74 MB/sec [4K blocks]
Uncached Read 520.48 261.59 MB/sec [256K blocks]
Random 177.80
Uncached Write 73.69 7.80 MB/sec [4K blocks]
Uncached Write 460.97 147.57 MB/sec [256K blocks]
Uncached Read 306.47 2.17 MB/sec [4K blocks]
Uncached Read 286.15 53.10 MB/sec [256K blocks]fernmeister
macrumors regular
I'm coming at this from the perspective of Digital Audio, where drive times are a real performance issue. Given that prices are coming down for SSD it makes sense to consider it because RAID doesn't work well in DAW (very unpopular) and there is lots of small seek (yes thousands of files) in sample playback.
Also, there is the issue of drive noise.
In fact, a Mac Pro with three SSDs (system, tape, samples) and a big drive for archival feels very good to me. In fact, I'll predict that by next year we'll see a lot of people going that way as the prices come down.
Tesselator
macrumors 601
We're seeing a lot of people doing that already. 
I'm not against SSD. Not at all. I just wanted to interject some actual facts and some common sense into the thread instead of fanatical untrue uber-hype.
SSDs have their place for sure! At some things like you're mentioning here with the DAW stuff for example, they may be very good indeed!
I'm not against SSD. Not at all. I just wanted to interject some actual facts and some common sense into the thread instead of fanatical untrue uber-hype.
SSDs have their place for sure! At some things like you're mentioning here with the DAW stuff for example, they may be very good indeed!
DAW's for sure for the SSDs!
I agree that SSDs compared to a RAID0 the RAID will win. However, it seems that nobody here (in this thread) has brought up the fact that SSDs are still an emerging technology. I know that they have been around for a while, but compared to HDDs they haven't. As time goes on they will only get cheaper, faster, bigger (capacity), and more reliable. Even now it takes three HDDs to get on the same level as two of those qualities for SSDS (price, speed, but not capacity).
There are certain situations where SSDs are great. Two of which have already been brought up. In older machines that are already maxed out(or have limited drive bays), or close, elsewhere and in DAW usage. In other words, if you only have one drive bay or want to do music, go SSD.
They may not be worth the negatives in most situations right now, but I certainly hope that the tech continues to get better and in 2 or 3 years time the quality/price issue of a single SSD versus a single HDD will be at parity or much closer.
I agree that SSDs compared to a RAID0 the RAID will win. However, it seems that nobody here (in this thread) has brought up the fact that SSDs are still an emerging technology. I know that they have been around for a while, but compared to HDDs they haven't. As time goes on they will only get cheaper, faster, bigger (capacity), and more reliable. Even now it takes three HDDs to get on the same level as two of those qualities for SSDS (price, speed, but not capacity).
There are certain situations where SSDs are great. Two of which have already been brought up. In older machines that are already maxed out(or have limited drive bays), or close, elsewhere and in DAW usage. In other words, if you only have one drive bay or want to do music, go SSD.
They may not be worth the negatives in most situations right now, but I certainly hope that the tech continues to get better and in 2 or 3 years time the quality/price issue of a single SSD versus a single HDD will be at parity or much closer.
Boomhowler
macrumors 6502
Stability
Remember that a raid0-system is very sensitive to failing disks.. It has X number extra disks so you really make it more possible to loose a drive and it only takes ONE failed drive to loose all data. Add to this that a SSD is less prone to fail since it has less moving parts.
SSD for laptops is hard to beat, especially that there aren't any non-alienware "laptops" that can have Raid with more than 2 disks.
Remember that a raid0-system is very sensitive to failing disks.. It has X number extra disks so you really make it more possible to loose a drive and it only takes ONE failed drive to loose all data. Add to this that a SSD is less prone to fail since it has less moving parts.
SSD for laptops is hard to beat, especially that there aren't any non-alienware "laptops" that can have Raid with more than 2 disks.
Loa
macrumors 68000
Thing is, SSDs are glorified CF cards. Even if there are no moving parts, I would never do a back-up on a CF card, while I wouldn't hesitate on a HD. Anyone who's doing any kind of serious digital photographic work will tell you that CF cards (and their smaller cousins) are not known for reliability.
As for a RAID being so very fragile: that's why we have back-ups. I wouldn't even use one HDD (or one SSD) without a back-up. Why would I use a RAID without one??? Backing-up is just like buckling-up in a car: if you value your health and life, you're going to buckle-up.
Even if you told me that a SSD was super super safe, I would still use a back-up or two. Would you stop using your car's belt-buckle after buying a very safe car? No.
Or, seeing this from another angle: if RAIDs were so unstable and unreliable, why would the technology be so widespread?
Saying that a RAID is not very reliable because there are many drives that can fail is like saying that a SSD is reliable because it has no moving parts. Without real world data to back such statements, it's nothing more than sophistry.
...you know, I just realized that very old dynamite also has no moving parts... Hmmm.
Loa
Boomhowler
macrumors 6502
..but dynamite is a bit liquid, right?
Yes, ufcourse you have backup. But with a small SSD you can have just one disk as the "fast, bootdisk" and you can use the rest of the three sleds in the mac pro for storage (amiing for some speed, but quantity mainly). With a 4-disc raid system in the mac you would only have either 1TB rather fast or .. well, you can have many configurations
As for longevity we will have to see, I guess that they will be pretty longlived (SSD) but since they have a limited amount of writes for each datablock they won't live forever either. Ordinary hard drives are more prone to malfunction because of dust and more mechanical issues which SSDs are not as sensitive to.
I think that for laptops it ain't a question which one is better, but for stationary computers you really have a point.
Yes, ufcourse you have backup. But with a small SSD you can have just one disk as the "fast, bootdisk" and you can use the rest of the three sleds in the mac pro for storage (amiing for some speed, but quantity mainly). With a 4-disc raid system in the mac you would only have either 1TB rather fast or .. well, you can have many configurations
As for longevity we will have to see, I guess that they will be pretty longlived (SSD) but since they have a limited amount of writes for each datablock they won't live forever either. Ordinary hard drives are more prone to malfunction because of dust and more mechanical issues which SSDs are not as sensitive to.
I think that for laptops it ain't a question which one is better, but for stationary computers you really have a point.
Loa
macrumors 68000
Hello,
If my budget allows for it, I'm going to try a SSD as a boot drive sometimes in early 2010. I'm still curious, but not as much as I was before.
But in the next month or so, I'll be installing a 4 disk RAID0 set in my tower. (2TB of very fast data for the price of one SSD)
If I do decide on a SSD after using the 4 disk RAID0 for a few months, I'll be able to see by myself if the difference is so noticeable!
Loa
If my budget allows for it, I'm going to try a SSD as a boot drive sometimes in early 2010. I'm still curious, but not as much as I was before.
But in the next month or so, I'll be installing a 4 disk RAID0 set in my tower. (2TB of very fast data for the price of one SSD)
If I do decide on a SSD after using the 4 disk RAID0 for a few months, I'll be able to see by myself if the difference is so noticeable!
Loa
I agree that anyone who criticizes the SSD without having used one is not qualified to do so. It's a totally different experience from any other upgrade I've ever done. I got rid of my SSD because it was causing sleep problems with my mac pro, but other than the original 3dFX video card release, I can't think of another upgrade (and I've done LOTS of them in 25 years of computer use) that has been as dramatic.
nuckinfutz
macrumors 603
The only qualms I have about SSD are pricing (which will improve quickly IMO) and lack of OS support. To data our storage systems assume that mass storage is spinning disc and controllers are designed with this assumption. SSD requires a new way of thinking in order to maximize the benefits.
Not only is low latency a killer feature of SSD but random read/writes clobber HDD discs. Not even a 15k SAS drive can match an Intel SSD here.
RAID 0 ain't redendant where you need it. The only redundant item you have is the drives themselves and striping them destroys any safety.
I think most people need about 80-120GB of storage at under $200 to make the plunge. The performance benefits are worth paying the premium for people that value a snappy computer.
JNB
macrumors 604
You realize you're saying then that subjective, unmeasured observation is a better method than objective, spec-for-spec comparison.
VirtualRain
macrumors 603
I stand by my assessment. I've been using computers for 30 years and for me, the move from magnetic HD's to SSD's is comparible in significance to the move from floppy disks to hard disks.
Sure, if I compared my Nehalem Mac Pro to my 800MHz PIII from 2001, I might change my tune and recommend that as a more significant upgrade... LOL... of course.
But "within" any given PC architecture I've ever owned (and I've owned all of them), I have not made an upgrade to my computer that has impacted its performance as much as adding SSD's has made. (EDIT: I should point out that on my last Core2 Quad system, I ran four Raptors on an Areca RAID card and they still don't measure up to a single SSD in terms of impact on performance).
But don't take my word for it... try it out for yourself.
Cheers!
iamcheerful
macrumors 6502
i agree with you on --- "I can't think of another upgrade (and I've done LOTS of them in 25 years of computer use) that has been as dramatic."
thus, i stated my experience with RAM Disk above. it really is day and night, normal booting vs RAM disk booting. too bad Apple couldn't bring this trick beyond those years. sigh ...
VirtualRain
macrumors 603
I think the problem is that the specs are not being weighted appropriately. People look at the STR of 3 or 4 drive RAID0 arrays and think that's all that matters and ignore the 100x fold improvement in access times. The fact is that STR is irrelevant when accessing a random set of small files... but access time is HUGE when it comes to percieved performance!
Correct.
There is a difference though. Both use the same flash chips, but SSD's have wear leveling functionality in the controller. It does help, but it can't change the fact the flash is only good for 1E4 or 1E5 writes. Only that the cells will be written to equally, so some aren't stressed (written to) more than the others. For high write usage, this doesn't change the fact an SSD will die faster than a mechanical drive.
So usage patterns should be one of the major determining factors. Good point.As for a RAID being so very fragile: that's why we have back-ups. I wouldn't even use one HDD (or one SSD) without a back-up. Why would I use a RAID without one??? Backing-up is just like buckling-up in a car: if you value your health and life, you're going to buckle-up.
Even if you told me that a SSD was super super safe, I would still use a back-up or two. Would you stop using your car's belt-buckle after buying a very safe car? No.
Or, seeing this from another angle: if RAIDs were so unstable and unreliable, why would the technology be so widespread?
Saying that a RAID is not very reliable because there are many drives that can fail is like saying that a SSD is reliable because it has no moving parts. Without real world data to back such statements, it's nothing more than sophistry.
...you know, I just realized that very old dynamite also has no moving parts... Hmmm.
Loa
RAID or single disks shouldn't be considered without a backup, ever. SSD has it's use in this type of situation:..but dynamite is a bit liquid, right?
Yes, of course you have backup. But with a small SSD you can have just one disk as the "fast, bootdisk" and you can use the rest of the three sleds in the mac pro for storage (amiing for some speed, but quantity mainly). With a 4-disc raid system in the mac you would only have either 1TB rather fast or .. well, you can have many configurations
1. Speed boost where only a single disk can be used.
2. Usage pattern will be high reads, not writes.
3. Budget isn't as much of a concern.
4. Current capacity will fullfill the need.
Nice if this balance is possible, but may not be possible for everyone or situation.
For reads, they should last awhile, but for high writes, they won't. It really will come down to the flash technology used. Under high write usage, we'll start to see a number of failures in the not too distant future.
A big improvement for sure, but you might recall the early HDD's weren't that reliable either, and needed time to mature.
I'm not saying it's not fast, but rather keep in mind the usage. For an OS disk, particularly a single, it's faster than a mechanical drive. Independent tests, and user experience aren't full of it.But "within" any given PC architecture I've ever owned (and I've owned all of them), I have not made an upgrade to my computer that has impacted its performance as much as adding SSD's has made. (EDIT: I should point out that on my last Core2 Quad system, I ran four Raptors on an Areca RAID card and they still don't measure up to a single SSD in terms of impact on performance).
But don't take my word for it... try it out for yourself.
Cheers!
Just that it's not to the point it can be generalized, as there's too high a disparagement between SSD's read and write cycle limits. So it's in the details of how it's used. For now. If the usage is for reads (OS for example), lower capacities, budgets,... aren't an issue, and a user wants the improved performance, then by all means; go for it.
nuckinfutz
macrumors 603
Intel rates their longevity figures based on writing 20GB per day
Their SLC SSD is rated at 2 million hours.
I'm sorry but I doubt that mechanical hard drives are truly as durable as a SSD.
From the X-25-E pdf
That's a **** ton of writing folks
Their SLC SSD is rated at 2 million hours.
I'm sorry but I doubt that mechanical hard drives are truly as durable as a SSD.
From the X-25-E pdf
That's a **** ton of writing folks
Keep in mind, the spec is listing Random Writes, which isn't the same as writing to the same cell (single cell write cycle limit).
So here's a really simple way to look at it mathematically. The result might surprise you.
64 PB/64GB = 1E15/1E9 = 1E6. That's it, and is the write cycle rate of a single cell. It also includes the wear leveling achieved by the drive's controller.
You get a whopping 10x improvement over the SLC's limit of 1E5 due to the wear leveling capability.
So it proves that they haven't beaten device physics.
But it is wonderful marketing. 
Catfish_Man
macrumors 68030
I dropped a 160GB X25-m in my machine. Very very nice upgrade. Searching (spotlight and in xcode) is immensely faster in particular.
nuckinfutz
macrumors 603
160GB ..nice. Once we get this size down to sub $300 it's off to the races.
I wonder what Intel's going to do for an encore. My guess is we'll see other vendors tweak their product and get closer to the X25-X series and then Intel will announce generation II with additional controller tweaks and other features and blow the competition away.
I don't think i've ever kept a drive in use much longer than 3 years so the larger these SSD go the less potential of running into wear level issues.