Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
anyone know if the x-25m speed will be crippled if it's paired with a high capacity HDD in a RAID 0 setup? (in a uMBP)

It will peform as fast as the slowest of the two (obviously the HDD) because the data is shared on both drives. You are really better off installing your programs on the Intel and just storing everything on the HDD.
 
Wait, so they are not releasing the 320 Gb verision now and no speed improvements?

What a disappointment...
 
What are you talking about? The IOPS have doubled and the latency on writes has decreased significantly. I don't get where the "no speed improvements" thing is coming from.

I think most people are only looking at the sequential write speeds, and therefore don't see a huge improvement.

It will be interesting to see some more benchmarks of the new X-25's though, as the non sequential writes may really push it ahead of the Vertex's etc.
 
I think most people are only looking at the sequential write speeds, and therefore don't see a huge improvement.

Capacity is also a big turn down. Intel could've really set themselves apart - not saying they already haven't - by releasing a 320GB version. No one on the market has that big of a capacity, nor that performance improvement.
 
What are you talking about? The IOPS have doubled and the latency on writes has decreased significantly. I don't get where the "no speed improvements" thing is coming from.

He probably meant the seq write speed. It is still at 70MBps. Some people transfer huge files more often than small files.

The IOPS random doubled on the write speed only. which is nice but normal users may not actually see any difference. I didn't see any difference between Intel and my Vertex in multitasking performance and Vertex has barely half of Intel's IOPS.

The reduce read and write latency might be more noticeable but we'll see how it goes once somebody get it in their hands.

I think most people are only looking at the sequential write speeds, and therefore don't see a huge improvement.

It will be interesting to see some more benchmarks of the new X-25's though, as the non sequential writes may really push it ahead of the Vertex's etc.

What are you talking about? Intel's random IOPS is the highest on the market, not even Vertex can barely match it. The newer version just makes it the leader in IOPS for several more months until 3rd generation SSDs comes into the market later in the year.

Capacity is also a big turn down. Intel could've really set themselves apart - not saying they already haven't - by releasing a 320GB version. No one on the market has that big of a capacity, nor that performance improvement.
320GB would land it in $900-1000 price range, I think Intel expecting a very small sale numbers for it so they rather just push 80GB and 160GB and drive the costs down much more before they can start pushing 320GB. 320GB for 500$ is much more tolerable, and they might hit that by the end of the year or early next year.
 
320GB would land it in $900-1000 price range, I think Intel expecting a very small sale numbers for it so they rather just push 80GB and 160GB and drive the costs down much more before they can start pushing 320GB. 320GB for 500$ is much more tolerable, and they might hit that by the end of the year or early next year.

I don't see a problem with Intel releasing the current line-up along side a 320GB version. And, I would've gladly paid $900 for a 320 SSD.

It's like Apple's Mac Pro; they target a different market with the higher price. Intel might have done the same. Its not as if no-one would've bought the 320 version - it's just that they would've sold less, at a higher price - which would've, theoretically, made them the same sales margins - roughly.



ps: I'm just disappointed that they didn't release a 320gb version - I was looking forward to purchasing one.
 
What are you talking about? Intel's random IOPS is the highest on the market, not even Vertex can barely match it. The newer version just makes it the leader in IOPS for several more months until 3rd generation SSDs comes into the market later in the year.

I meant for those that are only looking at the sequential write stat. Maybe people will pick up on the real world performance of the X-25 vs. the Vertex even more so now that the IOPS have increased.

That's all I was trying to say...
 
Its basically at the point where the intel ssd's are on par with the price of all the other manufacturer's ssd's.

I think someone would be making a big mistake to pay the same price for a non intel ssd.
 
I meant for those that are only looking at the sequential write stat. Maybe people will pick up on the real world performance of the X-25 vs. the Vertex even more so now that the IOPS have increased.

That's all I was trying to say...

I doubt that because the increased IOPS does not change anything, Intel's seq speed is still the same, Intel's random IOPS is still the highest. Comparing it to any other SSD doesn't change the picture for anybody except for the business guys who run highly intensive disk operations but those type of guys would be buying the E series not the M series.

That's what I was trying to say, people looking at either seq or random will see the same picture as before.

The only thing is the impact of reduced latency for both read and write, is it noticeable for anybody at all? Cutting 2ms of startup does not make any difference but cutting off 1-2 seconds might.
 
I'm glad Intel didn't buy into the marketing scheme that other companies like OCZ are using.
 
MikhailT-

I hear what you're saying. As someone who has been looking at SSDs really closely over the last few weeks (as a way to upgrade my Early 08 MBP), it's easy to see how people might be lured in with the high sequential writes of a Vertex for instance. As I delved into it deeper though, it truly is amazing how much faster the random writes happen on the X-25's.

I have been holding out for this update, so now I'm ready to get one... Once they are in stock that is... :)
 
MikhailT-

I hear what you're saying. As someone who has been looking at SSDs really closely over the last few weeks (as a way to upgrade my Early 08 MBP), it's easy to see how people might be lured in with the high sequential writes of a Vertex for instance. As I delved into it deeper though, it truly is amazing how much faster the random writes happen on the X-25's.

I have been holding out for this update, so now I'm ready to get one... Once they are in stock that is... :)

I have the Vertex, it's awesome except with some minor issues. If Intel was selling those drives now for those price, I wouldn't gotten Vertex in the first place.

You are going to be very happy with those new models.
 
I doubt that because the increased IOPS does not change anything, Intel's seq speed is still the same, Intel's random IOPS is still the highest. Comparing it to any other SSD doesn't change the picture for anybody except for the business guys who run highly intensive disk operations but those type of guys would be buying the E series not the M series.

That's what I was trying to say, people looking at either seq or random will see the same picture as before.

The only thing is the impact of reduced latency for both read and write, is it noticeable for anybody at all? Cutting 2ms of startup does not make any difference but cutting off 1-2 seconds might.

I think IOPS and random access are much more revealing about the performance of a drive than the seq speeds... When you launch applications and boot quite a bunch of files have to be located and loaded...
 
I think IOPS and random access are much more revealing about the performance of a drive than the seq speeds... When you launch applications and boot quite a bunch of files have to be located and loaded...


Majority of that is thanks to the latency, not random IOPS.

Yes, random IOPS, the higher the better. But there comes to a point where a super high random IOPS don't matter to a typical user.

I was trying to tell somebody that this new revision doesn't change the picture for Intel, they still have the highest random IOPS on the market for a consumer SSD and this revision doesn't change that fact, it also doesn't change the fact that it has a limited write seq speed.
 
Majority of that is thanks to the latency, not random IOPS.

Yes, random IOPS, the higher the better. But there comes to a point where a super high random IOPS don't matter to a typical user.

I was trying to tell somebody that this new revision doesn't change the picture for Intel, they still have the highest random IOPS on the market for a consumer SSD and this revision doesn't change. Nor does it change that Intel has the average write seq speed.

Two sides of the same coin I guess. I think they matter a bit more than seq. speeds since almost everyone's compure searches random files (his internet cache for example). For notebook users with a single drive sequental speeds change almost nothing since you are probably transfering data via a connection much slower than that of the SSD.

I expect similar behavior with two drives with similar IOPS and 70mB difference in speed. But if the slower drive has 3 times the IOPS of the faster one it will feel much faster and snappy.

Bottom line, the most significant thing about the release is that one of the fastest and most trouble free out-of-the-box SSD's of the market got about 30% cheaper.
 
Bottom line, the most significant thing about the release is that one of the fastest and most trouble free out-of-the-box SSD's of the market got about 30% cheaper.

Wasn't there a thread about the X-25Ms failing? I didn't read into it too much, but I do recall many members here talking about it.

Either way, I think I will pick one up with the new prices.
 
IOPS and lower latency are important to me not because I'm running Oracle but because when I move to quad core I really want to keep my Mac as busy as possible. Hopefully I can leverage Automator and other tools to do a lot of heavy lifting without my intervention.

SSD are great for multi-tasking according to Anand's benchmarks.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.