Next generation of SSDs to have a shorter lifespan?

You write 1TB a day to a 80GB drive? Doing what?

BTW They say it only can do sustained writes at 70MB/sec... so.....
>>> 1000000/70.0/60.0/60.0
3.9682539682539684

You write to a drive 4 hours nonstop/day?


Oh yeah heres formula Intel says to use, Write amplification and wear leveling are both <=1.1 they say and you get 10,000 cycles on MLC drive.

Cycles = (Host writes) * (Write amplification factor) * (Wear leveling factor) / (Drive capacity)


10,000 = X * 1.1 * 1.1 / 80000
So you can write 661TB to the drive or 360GB/day for 5 years?

Err...

Okay Intel G1 drives have 50nm MLC and are rated ~10k writes/cell, Intel G2 drives have 34nm MLC and are rated ~4k writes/cell. Really I think that above calculation is just BS, here is better appromation from Intel:

http://intelstudios.edgesuite.net/idf/2009/sf/aep/IDF_2009_MEMS003/f.htm

X25-M 80GB - 100% random host data 4k writes = 7.5TB
X25-M 160GB - 100% random host data 4k writes = 15TB
X25-M 160GB - 100% sequential writes = 370TB

You can purposely kill a 80GB drive in a few days with random 4K writes. You can also greatly extend the life by partitioning the drive smaller, using only 140GB or something of the 160GB drive you can extend the random 4K writes to 42TB. Thats almost 3x improvement in life. Also I think might of actually seen that presentation and IDF and totally forgot about it, heh.

Anyone have a Intel SSD also think they do alot of writes to it? Can you post its SMART data? Poweron hours & Host writes?
 
Last edited:
I can post a screenshot but I don't do alot of writes to my drives.

And i'm curious to know what the hell that guy is doing writing 1 TB to his SSD on a daily basis sounds pretty suspect to me :D
 
Makaveli@BETA said:
I can post a screenshot but I don't do alot of writes to my drives.

And i'm curious to know what the hell that guy is doing writing 1 TB to his SSD on a daily basis sounds pretty suspect to me :D
Click to expand...

I have a dozen VMs running on it.
On a few particular VMs, I do testing for FlexRAID (www.flexraid.com). ;)

Ps. also, you can also generate 1TB worth of write on an SSD that has a block size of 512KB with only 2MB worth of data written 1 byte at the time.
Although, this is extreme, the average user generates far more writes on an SSD than it is realize.
Simply updating the ID tags on an MP3 collection will generate an amount of writes that is at least equal to the size of the whole collection as the tag data is spread throughout each file.

My usage has in fact gone down from 2TB to 1TB by putting a few things on RAM drives and iSCSI targets.
 
Last edited:
Can you post smart data from your Intel SSD?

Here's Intel's Tool:
http://www.intel.com/go/ssdtoolbox/

Or dump SMART data from any program, it just will label it Unknown. I think its either E1 or E9 that we want, I think E9 actually, whatever the raw value is, in addition to the power on hours,
 
Last edited:
Are you gonna update the ID3 tags on your entire MP3 collection every day tho? :p I get what you're saying, and for your specific scenario involving VMs and whatnot these are all very valid concerns... But that's way outside the scope of the vast majority of users. For someone like you there'll always be SLC drives or if it's really an issue w/the next-gen drives then someone will pick up the slack and continue manufacturing 34nm drives (hopefully).
 
Hypernova said:
Problem is with retention so short it is not safe to use SSD for OS as there are many files that are static. In fact most would be outside of log and temp files so you will need to reinstall everything (or ghost and copy back) every few months to prevent the data evaporating.

The issue of retention was never raised before as they used to be in the decades range.
Click to expand...

I'm reading this thread top down, so this may have already been addressed, but static files is not the same as static blocks. SSD's will relocate even static blocks as part of the wear leveling routines. A perfect theoretical wear leveling routine would rewrite / relocate every block of flash at least once if you write an amount of data equal to the drives capacity. This would apply even if you were writing to a focused area and not spreading your writes evenly.
 
Last edited:
Allynm said:
SSD's will relocate even static blocks as part of the wear leveling routines.
Click to expand...

Yah, but that doesn't help unless it's powered up, which makes SSD's terrible for offline backup.


-------------------------------------------------------------------------

I got my Intel X25-E in. It feels about twice as fast as my old 7200rpm 640GB WDD drive, but not insanely faster. What is noticeably faster is doing scans with CCleaner, Malwarebytes, etc. MS Office opens up snappier too.
 
Well, if SSDs would also have a small battery backup unit that could last for 2 hours of operation, it would both have a safe write-back (big DRAM buffercache for performance; first 1GB write goes at interface speed).

Also, it would allow a disconnected SSD to power up once every day or month, to maintain its data integrity. Depending on how long this takes, it could power down again and sleep for another x days and use very little power so the battery can sustain a disconnected SSD for many years.
 
Why would you use a SSD for offline backups? Doesn't make any sense, the price/capacity ratio is not gonna match HDDs anytime soon. :rolleyes:
 
Impulse said:
Why would you use a SSD for offline backups? Doesn't make any sense, the price/capacity ratio is not gonna match HDDs anytime soon. :rolleyes:
Click to expand...

A more likely situation is someone putting a system with SSD in storage for a year or two while abroad or so, then to return to a fully wiped system.
 
Yeah, that's somewhat more likely... But again, you'd think anyone putting a system away for a year would back it up fully before doing so. /shrug
 
Impulse said:
Yeah, that's somewhat more likely... But again, you'd think anyone putting a system away for a year would back it up fully before doing so. /shrug
Click to expand...

You'd think so, yeah. How often do you back up again? ;)
 
Me? I backup my files multiple times a week and I backup my OS drive every two weeks... All done automatically to my external drive. With my netbook I leave myself a reminder to backup the OS config every couple of weeks but I have a 16GB SDHC card permanently inside it unto which it will automatically backup documents.

I've also got a couple HD off-site that I pick up every couple of months to update w/my important docs and music. I've lost enough in the past, now I don't mess around. /shrug
 
Impulse said:
Me? I backup my files multiple times a week and I backup my OS drive every two weeks... All done automatically to my external drive. With my netbook I leave myself a reminder to backup the OS config every couple of weeks but I have a 16GB SDHC card permanently inside it unto which it will automatically backup documents.

I've also got a couple HD off-site that I pick up every couple of months to update w/my important docs and music. I've lost enough in the past, now I don't mess around. /shrug
Click to expand...

That makes you one of the exceptions, I'm guessing :)
 
How much longer do we have to wait for the G3 from Intel to be available?

And what's the latest on SSD size of the Intel G3?
 
Impulse said:
Me? I backup my files multiple times a week and I backup my OS drive every two weeks... All done automatically to my external drive. With my netbook I leave myself a reminder to backup the OS config every couple of weeks but I have a 16GB SDHC card permanently inside it unto which it will automatically backup documents.

I've also got a couple HD off-site that I pick up every couple of months to update w/my important docs and music. I've lost enough in the past, now I don't mess around. /shrug
Click to expand...

Didn't see this post, but I do much of the same. Learned the hard way by having a painful crash that took a lot of data.
 
Impulse said:
Me? I backup my files multiple times a week and I backup my OS drive every two weeks... All done automatically to my external drive. With my netbook I leave myself a reminder to backup the OS config every couple of weeks but I have a 16GB SDHC card permanently inside it unto which it will automatically backup documents.

I've also got a couple HD off-site that I pick up every couple of months to update w/my important docs and music. I've lost enough in the past, now I don't mess around. /shrug
Click to expand...

Out of curiosity, what program do you use for backup? (1:1 backup? Or image based?) And what HDD brand/size?

As funny as it may sound, but of all the system that I have right now, the tiny and underpowered HP x485 mediaserver is the one that I value the most... It has 10TB of storage space and it is past half full with all the things that I archive on there. (Before anyone makes the joke, NO there are no pr0n material anywhere, on any of my system actually... I tend to hoard documentaries, good TV series, LPs, and back them up on HDDs.)

Initially I thought maybe SSDs would be great (once the cost comes down) for backup. But the more I research into them, the more disappointed with their long term reliability.

Seems that SSDs are being designed towards catering to the popular mentality that "reliability doesn't sell, but flash and bloated (and often buggy) features do."

For that reason, I have stuck with Intel SSDs (I have multiple G1s and G2s, all MLC though :( ).
 
Hypernova said:
That's the insane part about these new MLC's. The cells holding the charges are so small that the electrons are leaking out faster than ever. If you don't refresh your drive on a 6 month cycle you may see files rarely written to like MBR's and OS files evaporate on you.
Click to expand...

I assume a good controller design will keep track of the age of each block and refresh them for you in the background. Probably not a good idea to use these drives as an archival backup medium, but who would waste an SSD on that anyway.
 
Valnar said:
Well, to each their own. But I can move 20GB of data around a day without blinking.
Click to expand...

I think you would be suprised how few writes are done to hdd.
My 40 gig SSD has 700+ hours of power on and 1,33 TB of data written after almost 6 months of usage.
 
1,33 TB of data written after almost 6 months of usage
Click to expand...

I have an Intel 80GB G1 with 13.66TB written in 12 months.

I'm hoping it goes for awhile longer! :)
 
Metaluna said:
I assume a good controller design will keep track of the age of each block and refresh them for you in the background. Probably not a good idea to use these drives as an archival backup medium, but who would waste an SSD on that anyway.
Click to expand...

Odds are they do, but doesn't change the fact that a powered off drive is going to kill itself. So if you got an old SSD a few years down the road that you want to pull data out of they may be gone.
 
Hypernova said:
Odds are they do, but doesn't change the fact that a powered off drive is going to kill itself. So if you got an old SSD a few years down the road that you want to pull data out of they may be gone.
Click to expand...

That's rather virtual problem for now.
Next gen SSDs still won't be cheap enough to use them as dedicated storage for large amounts of data.

Just imho but i think next few years will be dominated by setups of bigger SSDs with something like 1.5-2 TB slow rpm eco drive for storage.
 
600GB? My twin 160GB SSD's coming in my laptop with Fedex are already cowering in fear and shame. :(

Then again, a year ago, we were told the 34nm SSD's were going to be available in 320gb, so mabye that just means we'll finally just see those with 25nm (along with 1700 write cycles/cell, yay!*)



*guesstimate
 
A 600GB with 900 write cycles per cell will last longer than a 40GB drive with 10.000 write cycles per cell. So judging from that the real question is high fast the total number of write cycles grows; thus flash cells * write endurance per cell = total write cycles ("endurance").
 
bexamous said:
You write 1TB a day to a 80GB drive? Doing what?

BTW They say it only can do sustained writes at 70MB/sec... so.....
>>> 1000000/70.0/60.0/60.0
3.9682539682539684

You write to a drive 4 hours nonstop/day?


Oh yeah heres formula Intel says to use, Write amplification and wear leveling are both <=1.1 they say and you get 10,000 cycles on MLC drive.

Cycles = (Host writes) * (Write amplification factor) * (Wear leveling factor) / (Drive capacity)


10,000 = X * 1.1 * 1.1 / 80000
So you can write 661TB to the drive or 360GB/day for 5 years?

Err...
Click to expand...

I transfer data from Point A to my home's computer from midnight till 6 am, but not every night. Some week, it could be 50MB / min. for 6 hr. straight daily, some other night could be 1 hr. at 50MB/ min.

So if these new Intel G3 drive can't handle it, then I have to ask what I would pay so much more $$ than a cheap 1 TB HD?

Currently, I have a stack of 50GB Blu Ray back up, plus many DVD+R DL backup.
 
sub.mesa said:
A 600GB with 900 write cycles per cell will last longer than a 40GB drive with 10.000 write cycles per cell. So judging from that the real question is high fast the total number of write cycles grows; thus flash cells * write endurance per cell = total write cycles ("endurance").
Click to expand...

Hello naive, my name's okashira :D I keed I keed, but if you want to make a realilistic comparison, compare the Intel 34nm X-25M @ 320GB with 3000 write cycles to the X-25M @ 600GB with 900 write cycles.

I understand your points, however I have already done the analysis on this, and the capacity * endurance cycle factor, so-to-speak, has been going down drastically. Even if you factor in costs.

X-25M ~50nm 160GB SSD (approx $850 before 34nm release) 10000 write cycles 1.2 write amplification =
  • approx 1.25 petabytes endurance

X-25M 25nm 320GB SSD (approx $600 retail, estimate) 1200 write cycles 1.2 write amp
  • approx 0.31 petabytes endurance

old school X-25M 2008 model: $680 per petabyte of endurance
new school X-25M 2010 model: $1935 per petabyte of endurance

Suddenly, it doesn't look as good...

Don't forget also that there's a claimed built in safety factor on the 50nm models as much as "3x-4x" which means as many as 40000 write cycles. My intuition tells me this safety factor will be smaller with the 34nm and even smaller with the 25nm drives.

Look, I'm very much pro-SSD, I am just making a point. In fact, it is starting to look like 34 and 25 nm may be the sweet spot, with 50nm still having applications for very high endurance requirements.

(by the way, I don't know why I keep saying X-25M. the 1.8" form factor (X-18M) are far better and more valuable and useful. I'll only buy 1.8" drives from now. They are much better for laptops. My laptop (in the mail...) can run 2x 1.8" drives or 1x 2.5" drive. Guess what i'll be running.) my hard drive purchases will also be either 1.8" (to fit in laptop) OR 3.5" (for storage) No 2.5" here.
 
Hypernova said:
Nice write up on the recent International Solid-State Circuits Conference (ISSCC)
http://techon.nikkeibp.co.jp/article/HONSHI/20100420/181992/?P=1

Hopefully by the time flash retention starts becoming critical they won't matter any more.
Click to expand...

Of note is that Samsung is already shipping PCM modules for use in mobile phones and such, where it will replace NOR Flash. From what I have read PCM seems likely to take the world by storm the coming years. MRAM still has scaling issues and such. Never heard of that other memory tech they mentioned in the article.
 
okashira said:
Hello naive, my name's okashira :D I keed I keed, but if you want to make a realilistic comparison, compare the Intel 34nm X-25M @ 320GB with 3000 write cycles to the X-25M @ 600GB with 900 write cycles.
Click to expand...
Hello omnipotent, i'm sorry to disturb you with my foolish naive questions, but perhaps you have the time to help a humble pathetic creature such as myself.

The 900 write cycle was of course just an extreme example. If we make a more reasonable comparison we would get this:

55nm generation
160GB * 10.000 cycles = 1600 TB total writes
34nm generation
160GB * 4.000 cycles = 640 TB total writes
25nm generation
600GB * 2.800 cycles = 1680 TB total writes

The write cycle count has to be as low as ~1000 cycles to get on the old level of the 34nm generation sold currently.

Suddenly, it doesn't look as good...
Click to expand...
Looks even worse if you give it just 1 write cycle; so perhaps you'd like to change your calculation to show an even more unrealistic picture.

Look, I'm very much pro-SSD, I am just making a point.
Click to expand...
The only point you should make is that we don't know the exact write endurance since we have no information about the maximum number of write cycles on the not-yet produced 25nm Micron NAND process.

Instead you spread FUD based on assumptions, that basically SSDs will be useless when the current products are not sold anymore. And if somebody tells you a counter-argument you say he is naive? Nice forum.
 
sub.mesa said:
Hello omnipotent, i'm sorry to disturb you with my foolish naive questions, but perhaps you have the time to help a humble pathetic creature such as myself.

The 900 write cycle was of course just an extreme example. If we make a more reasonable comparison we would get this:

55nm generation
160GB * 10.000 cycles = 1600 TB total writes
34nm generation
160GB * 4.000 cycles = 640 TB total writes
25nm generation
600GB * 2.800 cycles = 1680 TB total writes

The write cycle count has to be as low as ~1000 cycles to get on the old level of the 34nm generation sold currently.


Looks even worse if you give it just 1 write cycle; so perhaps you'd like to change your calculation to show an even more unrealistic picture.


The only point you should make is that we don't know the exact write endurance since we have no information about the maximum number of write cycles on the not-yet produced 25nm Micron NAND process.

Instead you spread FUD based on assumptions, that basically SSDs will be useless when the current products are not sold anymore. And if somebody tells you a counter-argument you say he is naive? Nice forum.
Click to expand...

FUD? Wow... do you know what that means? Let's lose the pointless attitude and attacks, OK? I was here only to present my research, unbiased where possible to help others make decisions.

This time without the light / joking tone, I'll just ask directly. Why are you directly comparing the 160GB 34nm drive to a 600GB 25nm drive without accounting for the important (and stark) differences?

1.) Intel is not likely to jump a size gradient like that. The largest size will probably be the 320GB/300GB size. Intel didn't even release the 320GB 34nm drive 'like they said they would.

2.) You are comparing a $400 SSD (160GB 34nm) with an approximately $1100 (600GB 25nm) SSD.

3.) You need to also acknowledge that a larger and faster drive will inherently be subjected to more writes.

4.) It's not necessary to "KNOW" these things to make a comparison or argument. Natural phenomena typically results in smooth curves when you plot one dependent variable against some independent variable, with relatively few exceptions, of which those are simply discontinuities or explained by an additional independent variable. This is a widely accepted aspect of science and is the basis for the concept of reliable extrapolation. Using similar techniques of manufacture & materials (as is the case in 34nm to 25nm FLASH memory) this is a perfect example. The shrink from 34nm to 25nm is about the same relative magnitude as the shrink from 50nm to 34nm.
  • From 50 to 34 we went from 10,000 to 3,500 cycles. Applying this pattern, we will get 3500/10000 * (25/34)/(34/50) * 3500 = ~1325 cycles for 34 to 25nm. (Don't ask me to work out the units)
The underlying physical concepts may be better represented by squaring the process size, rather it's some function of the size of the "barrier" that holds the charge. In any case, I would not feel much less confident about this estimate then what intel officially publishes when they release their 25nm drives.

Again, to reiterate some predictions...

G2: 160gb * 3500 cycle * (1/1.2) write amplification = .45 PB endurance
$400 / .45 PB = 899 $/PB

G3: 320gb * 1325 cycle * 1/1.2 write amplification = .337 PB endurance
$550 (generous est. for cost of $320GB drive) / .337 = 1632 $/PB


Intel G2, mid 2009:
899 $/PB Endurance
2.5 $/GB Storage

Intel G3, late 2010:
1632 $/PB Endurance
$1.70 $/GB Storage


Also keep in mind that a larger (and faster) drive will inherently be subjected to more writes, reducing lifetime/perceived endurance (on a cost basis) What I mean is yes, the larger drive has more raw data endurance, but it requires proportionally more writes to "use" it and proportionally more $ to buy it. I am not saying one's better then the other, that's for the buyer to decide. Honestly, a G3 with these specs will be just fine with me based on my current usage of my 80GB G2 and the time I've owned it. Calling a post based on facts, with some pretty unbiased speculation (IMO) FUD is worse then FUD itself... But please someone let me know if what I posted is biased (or FUD...!). If you don't agree with my assumptions or predictions, just say so and/or plug in your own values.

There is nothing wrong with making assumptions as long as you try to acknowledge their limitations. I will acknowledge there is a lot of uncertainty in my estimates ... My point is here is that it's also worth acknowledging the very likely trend here, that relative and cost based SSD endurance is shrinking with each new generation.
 
okashira said:
FUD? Wow... do you know what that means? Let's lose the pointless attitude and attacks, OK?
Click to expand...
Fine, and i'm sorry. But i wasn't the one to start with personal attacks. I'd love to adhere to the technical reasoning without getting such comments. I hope we can just continue in a more civilized manner and stop being offensive?

This time without the light / joking tone, I'll just ask directly. Why are you directly comparing the 160GB 34nm drive to a 600GB 25nm drive without accounting for the important (and stark) differences?
Click to expand...
160GB is the biggest X25-M G2 you can buy now, 600GB will be the biggest X25-M G3 you will be able to buy around christmas this year, at least according to several news sites including Anandtech:
http://www.anandtech.com/show/2928

Ultimately, all pricing are dependent on production volume. Produce one thing of something and the cost is 100 million. Produce 100 million of something and the cost is 1000. Produce 100 billion and the cost is 100. Not that these values are accurate in any way, just to make the point here.

If Intel/Micron increases NAND production as they are planning, that leads to more products being sold, either via retail but also via OEM channels. Think netbooks with only SSDs produced in mass volume. Once a 'turning point' is made, the NAND market can grow very rapidly. In that scenario, it would not only be Intel/Micron producing at large scale, but other foundries would be keen on investing in this growing market with great revenues; the demand would be higher than the supply for years to come. This would warrant huge investments into high-tech fabs, even more than today.

Essentially, the HDD market is being (partially) absorbed into NAND market. The benefit here is that NAND products will mature rapidly and come down in cost.

At least that's the theory - with NAND prices not dropping the only real price cut we got was when going from 55nm to 34nm; the Intel G2 was significantly cheaper. A same price-cut will happen with 25nm products. Also, it could be possible that Intel would choose an agressive pricing plan, relately even lower than when extrapolating the price cut to 25nm. The advantage there would be for Intel to seize a much higher market share because they are the first to really ramp up their NAND production. With the higher market share they would have an edge a few years later when more foundries begin ramping up 2xnm 2-bit MLC.

Of course, this is all speculation. But i understand you're not against that all all. If some of it is true, then pricing could drop over time more than we've seen thus far.

1.) Intel is not likely to jump a size gradient like that. The largest size will probably be the 320GB/300GB size. Intel didn't even release the 320GB 34nm drive 'like they said they would.
Click to expand...
Probably useless as even the 5-channel X25-V is already at the limits of SATA/300. Once their 6Gbps controller is ready to ship with the 25nm NAND it would make a 600GB SSD much more feasible.

If they had released the 320GB G2, probably it would not be sold alot and thus not profitable as another production line would increase operating cost and thus it might not be profitable at all. With the 80GB G2 having chips on one side, 160GB G2 on both sides, the 320GB would need either multiple PCB or double capacity NAND. So i think Intel made the choice to skip the 320GB for now due to their own wallet.

2.) You are comparing a $400 SSD (160GB 34nm) with an approximately $1100 (600GB 25nm) SSD.
Click to expand...
I'm not so interested in comparing prices - the topic was 'next generation SSDs to have shorter lifespan'. In that context, you could say the next generation SSDs to have both shorter lifespan and higher lifespan than the current generation, depending on the size of disks you choose.

But we can see all capacities double:
Intel X25-V 40GB goes to X25-V 80GB
Intel X25-M 80GB goes to X25-M 160GB
Intel X25-M 160GB goes to X25-M 300GB

I would think after intoduction price, the pricing would settle below their older 34nm cousin. So a 300GB 25nm Intel would be cheaper than the 160GB 34nm. Not straight away though, it took its time before G2 actually became available and pricing dropped due to more actual competition instead of competition on paper due to low stocks.

If this is at least partly true, you would get the 600GB for a very reasonable price. Would be the first real Gaming SSD i think.

3.) You need to also acknowledge that a larger and faster drive will inherently be subjected to more writes.
Click to expand...
True - i wasn't taking that into account with my calculations. Though, this does not have to scale with the size of the drive on a 1:1 basis. For example, if you have a 600GB SSD as system disk (excellent for Gamers) you would probably not try hard to disable the pagefile, hibernation file, system restore, indexing and whatever not to save space on the 'tiny' system drive.

As games would not write randomly alot (mostly sequential writes during install) it also wouldn't wear a drive as much as the random writes issued by the Operating System.

In other words, i think the new generation will work excellent for the purpose they were designed for. If you need more write endurance, then you reserve more space. Though not official, rumors exist that the new X25-E "extreme" generation will have up to 50% capacity reserved for internal use, but now uses 25nm MLC instead of SLC process.

If that is not enough, you can simply RAID0 or several disks to get more write endurance. This may be useful in database servers which see a high number of random writes. With each disk added, you increase the total write endurance.

It's not necessary to "KNOW" these things to make a comparison or argument.
Click to expand...
Yes, you can extrapolate. But that carries the danger of being naive (how ironic; isn't that?) - progress is more than linear increase of 'raw power' like more 'mhz' or more 'rpm' or more 'cells' - it can also mean progress by doing things a little differently. That can mean a little trick to make NAND more write endurable, some intelligence in the firmware that causes lower write amplification or even a new fundamental way to actually store bits - such as with Phase-Change memory. The consumer probably wouldn't notice; its just an SSD with Serial ATA like before. I can still see memory chips - how they work is actually very different.

A good counterargument would be that this is actually cheating; i'm not comparing the same product but actually one technology with another piece of technology. But hey that's exactly what has happened when we look at history. We don't have single-core CPUs anymore, we don't scale linearly with MHz anymore, actually MHz have become much less important.

There is nothing wrong with making assumptions as long as you try to acknowledge their limitations. I will acknowledge there is a lot of uncertainty in my estimates ... My point is here is that it's also worth acknowledging the very likely trend here, that relative and cost based SSD endurance is shrinking with each new generation.
Click to expand...
Point well made. For the moment, i think you're right though i feel it's also panic for nothing - some posts in this thread could be read as "we can't have reliable/usable SSDs anymore in the future" - which is what i meant with "FUD". Things don't look that bad and honestly some technology adjustment will come and fix our write endurance problem forever. However, it may also introduce new weaknesses inherent to that technology. Looks like this can be a long thread. ;-)
 
Thanks for keeping it civil, fellas. This is an interesting thread; would hate to see it locked. ;)
 
sub.mesa said:
At least that's the theory - with NAND prices not dropping the only real price cut we got was when going from 55nm to 34nm; the Intel G2 was significantly cheaper. A same price-cut will happen with 25nm products.
Of course, this is all speculation. But i understand you're not against that all all. If some of it is true, then pricing could drop over time more than we've seen thus far.
Click to expand...

When the X25-M G2s came out at the lower prices, the price cut was market driven since the X25-M G1s were more expensive per GB than the competing drives at the time.

Now, the G2s are actually a bit cheaper per GB than the leading performance drives (Sandforce and RealSSD).

So, with little market pressure, I bet that there will not be a big drop in $/GB for the G3s this year. Maybe next year when the competition starts catching up to IMFT.


sub.mesa said:
If they had released the 320GB G2, probably it would not be sold alot and thus not profitable as another production line would increase operating cost and thus it might not be profitable at all. With the 80GB G2 having chips on one side, 160GB G2 on both sides, the 320GB would need either multiple PCB or double capacity NAND.
Click to expand...

Actually, the X25-M 160GB G2 has only 10 NAND flash chips on one side of the board (you can find some pictures on Anand's site if you search). There are actually silkscreened blank spaces for 10 more chips on the other side of the board. There was no technical reason for Intel to not release a 320GB G2. It was a marketing decision.
 
john4200 said:
When the X25-M G2s came out at the lower prices, the price cut was market driven since the X25-M G1s were more expensive per GB than the competing drives at the time.
Click to expand...
But also much, much faster than those JMicron and early Indilinx firmware junk. Samsung joined in; another crap controller that only focuses on MB/s; getting it sold is more important than actually being fast.

I think Intel's price point is very low given the quality of this controller. If you look at what a normal gigabit PCIe Intel NIC costs and compare with reasonable Realtek alternative there is a huge difference in pricing. On the SSD market, Intel has paired a good controller with very sharp pricing. I think they will continue this path and ramp up their production to get a large market share now that the market is still thin-populated. You can see Sandforce coming and oh finally we got some competition and Intel being exceeded in some random I/O benchmarks (not all). Other companies with advanced controller designs can pop up any moment and have a variety of retail brands like OCZ to sell their chips to.

If Intel is quick enough to stay ahead, it will make sure it has got a large market share and profitable margins in the future. That would be an argument for seeing reasonable price drops on Intel drives in 2011; as Intel being the market leader other vendors will follow.

So, with little market pressure, I bet that there will not be a big drop in $/GB for the G3s this year. Maybe next year when the competition starts catching up to IMFT.
Click to expand...
Likely true as when the next-gen Intels come out they will be hot-sold keeping pricing high. Once the production volume grows and supply-demand become more balanced the prices will drop; we've seen that with the G2 also. In the first months you were barely able to get them, keeping initial retail pricing high.

Actually, the X25-M 160GB G2 has only 10 NAND flash chips on one side of the board
Click to expand...
Thanks for clearing that up. Seems like they already have doubled NAND capacity; and could have easily done a 320GB. But again: would you buy it? I would be very interested in the pricing of the 600GB G3; it would be the market for (spoiled) Gamers looking for an SSD. That also means they have to stay under $1000; at least after prices settle.

Intel could very well 'conroe' the SSD market again with the G3 that really uses the SATA/600 (speculation) and have much higher random IOps also (20 channels? more DRAM?).

Personally, i would love them to stick 1GB DRAM on an SSD as write-back buffer; causing wire-speeds writes for at least the 1GB and as long the SSD can keep up. Add in some capacitors to protect the write-back and you got hardware RAID quality write-back performance and security. So far i haven't seen such designs.
 
sub.mesa said:
Personally, i would love them to stick 1GB DRAM on an SSD as write-back buffer; causing wire-speeds writes for at least the 1GB and as long the SSD can keep up. Add in some capacitors to protect the write-back and you got hardware RAID quality write-back performance and security. So far i haven't seen such designs.
Click to expand...

Maybe some enterprise SSDs will take that approach, but I do not think you will ever see that for consumer SSDs:

1) Super capacitors are expensive

2) DRAM adds to the power consumption and heat production -- such an SSD would probably fare poorly for a laptop or netbook when compared with a 2.5" HDD.
 
You must log in or register to reply here.
Back
Top