OWC Mercury Extreme Pro SSD Problem

[Sir Atilla]

I went through the process over the weekend of updating from 3.43 to 3.60 but thus far it doesn't seem to resolve the issue for me although the behavior is somewhat different than before.

Originally on 3.43 machine would go sleep (blinking white light) then I could not enter my password once I would wake up machine.

Now with 3.60 what is strange is seems like machine never fully goes to sleep. I set the screen to go off at 1 minute and machine to sleep at 2 minutes. Screen goes off an a little bit later white light comes on but doesn't start flashing. Hit trackpad and type in password and machine will be okay for a few seconds before essentially becoming frozen again. Very strange. This will happen even if I book off a mechanical hard drive externally if the SSD is still connected internally. I know some have said there is a V4 firmware coming but I think the OWC guys are referring the MP4 in the filename of the 3.60 update. This is speculation on my part.

Would really be nice to get this issue resolved once and for all.

 

[Weaselboy]

Not sure if anyone has seen this yet.

A very interesting read, OWC really do make the best SSD's.

Here is a response from OCZ on this issue.

 

[RainCityMacFan]

Here is a response from OCZ on this issue.

I've been reading that for a bit now and honestly, I don't know who to believe. Sure OCZ has done sleazy things in the past, but they do have some evidence from SpecTek to back it up. But I doubt that OWC would publish something that could be considered falsed and libelous.

With little or no information on the difference between grades of SSD or official word from the NAND manufacturers due to NDAs and etc, there's no way of really clarifying the issue.

But according to a French retailer, OCZ does have the highest failure rate out of all other brand of SSDs excluding OWC and some others (due to the lack of information).

 

[Guy Mancuso]

I agree weird info. from both of them and question comes up is why on both counts. Certainly adds to confusion for us consumers.

In reality it is kind of meaningless in the end since now we have Sata 3 at our disposal for the MBP and neither one of them have made it to market. The real question comes into play for us now which one Intel, Vertex 3 , OWC. I honestly don't have much faith in anyone else maybe Crucial we can throw in the pile as well. For me it is not a matter of speed per say it is a matter of reliability in the long term. I think most of us can agree with that even though we are still throttle junkies. I mean been running the OWC Pro for a long time here and it has been faithful and healthy but I certainly paid for that as well. The real question is these are not cheap for most people and we want something that protects our data and performs at top speeds without issue.

Right now the Intel and older Crucial drives with Sata 3 are a mixed bag of working or not. Like a little guarantee here if i give someone my 600 dollars i want it to work on my MBP. I'm just not feeling that love right now and my finger is hovering over a buy button. But which one?????

These guys are not helping us.

 

[Weaselboy]

I've been reading that for a bit now and honestly, I don't know who to believe. Sure OCZ has done sleazy things in the past, but they do have some evidence from SpecTek to back it up. But I doubt that OWC would publish something that could be considered falsed and libelous.

With little or no information on the difference between grades of SSD or official word from the NAND manufacturers due to NDAs and etc, there's no way of really clarifying the issue.

But according to a French retailer, OCZ does have the highest failure rate out of all other brand of SSDs excluding OWC and some others (due to the lack of information).

I agree it is very confusing at this point. I am just more than a little skeptical of OCZ at this point because of the 34>25nm incident, and now this. I understand OCZ's explanation and it makes sense, but it too is worded carefully. I can't claim to have any expertise at all about the NAND industry, but if a chip meets spec why would Micron push it off to be sold by a subsidiary (implying they don't want their name on it). Perhaps it meets the spec the vendor (OCZ) specified, but not Micron's more stringent spec?

 

[timyu]

I agree it is very confusing at this point. I am just more than a little skeptical of OCZ at this point because of the 34>25nm incident, and now this. I understand OCZ's explanation and it makes sense, but it too is worded carefully. I can't claim to have any expertise at all about the NAND industry, but if a chip meets spec why would Micron push it off to be sold by a subsidiary (implying they don't want their name on it). Perhaps it meets the spec the vendor (OCZ) specified, but not Micron's more stringent spec?

I read a bit of the linked thread and this is what I understand from it (see especially post #13, from an OCZ rep):

1. OCZ contracts with a variety of suppliers for its flash memory requirements, and these suppliers change over time depending on availability (and OCZ business agreements).
2. One of its suppliers has been Spectek, a subsidiary of Micron.
3. Micron makes flash NAND chips. They are produced on wafers, each of which has many individual chips.
4. Each entire wafer is "spot tested" for failures. Wafers that contain defects are relegated to a lower tier, and sold to Spectek.
5. Spectek does more intensive testing of each chip to ensure that it meets specifications, and sorts the good from the bad.

So: OWC is correct in claiming that OCZ is using "second tier" memory, in that they came from "bad wafers". OCZ of course cries foul, and responds that it only uses the highest quality chips (that meet or exceed manufacturer's specifications) from these "bad wafers."

Who wins in the end? I have no idea, but I guess it would depend on how reliable the testing processes are for wafers and chips. The OCZ approach seems to smack a little bit of "refurbished" to me, and seems a little deceptive, but to be fair I don't know what industry standards are for "component recovery" processes like this.

 

[endless17]

oh and add member 'endless17' to that as well! Missed him/her/bot

this may be shocking, but I only started posting in MR after buying my 2011 MBP a few weeks prior. as to my posting history on this board, well, it's quite extensive and travels far beyond this idiotic thread.

you're quite funny - apparently pulling the curtain away from the smoke and mirrors that makes OWC SSDs 'better' than similar Sandforce-based SSDs is anathema to this thread? how quaint. perhaps i'll make up for it by purchasing their likewise overpriced rebranded MBP RAM modules that run at slower speeds than competing options.

timyu: from an article on Anandtech (I forget where), IBM has the best reliability for their SSDs by far, in part due to extensive testing. OCZ appears the middle of the pack, but with promises to improve -

http://www.anandtech.com/show/4202/the-intel-ssd-510-review/3

 

[AndrewLupin]

115gb???is it a step to 25nm???as ocz???

that's not a good thing...

 

[Guy Mancuso]

I read a bit of the linked thread and this is what I understand from it (see especially post #13, from an OCZ rep):

1. OCZ contracts with a variety of suppliers for its flash memory requirements, and these suppliers change over time depending on availability (and OCZ business agreements).
2. One of its suppliers has been Spectek, a subsidiary of Micron.
3. Micron makes flash NAND chips. They are produced on wafers, each of which has many individual chips.
4. Each entire wafer is "spot tested" for failures. Wafers that contain defects are relegated to a lower tier, and sold to Spectek.
5. Spectek does more intensive testing of each chip to ensure that it meets specifications, and sorts the good from the bad.

So: OWC is correct in claiming that OCZ is using "second tier" memory, in that they came from "bad wafers". OCZ of course cries foul, and responds that it only uses the highest quality chips (that meet or exceed manufacturer's specifications) from these "bad wafers."

Who wins in the end? I have no idea, but I guess it would depend on how reliable the testing processes are for wafers and chips. The OCZ approach seems to smack a little bit of "refurbished" to me, and seems a little deceptive, but to be fair I don't know what industry standards are for "component recovery" processes like this.

Number 4 is kind of the give away. Oh well not sure i can wait anyway and may get the Intel since it is out.

 

[Guy Mancuso]

115gb???is it a step to 25nm???as ocz???

that's not a good thing...

Yes and the 480 from Lloyds blog i saw this yesterday. These are still Sata II though just FYI.

And i quote

Now listed on the OWC site is the new 25nm based 480GB SSD for about $920 for the 480GB OWC Mercury Extreme Pro SSD as I write this. Shipping is noted as “5 Days”. This 480GB drive is what will ship at the high-capacity end, up from the 400GB as noted a few days ago.

The 480GB has 32GB of over-provisioning (7%) using Tier 1 flash memory chips, for long term reliability. and is a SATA II 3Gbps drive. See the other technical specifications on the OWC info page.

OWC is also listing the new 25nm-based 115GB model for $229.99, with availability “5 days”.

Be sure to see yesterday’s notes on the changing SSD landscape, especially the notes about coming 6Gbps SATA III drives.

 

[RainCityMacFan]

At least OWC is advertising the changes on their website rather than marketing it as 120GB.

But I don't think there will be noticeable performance degradation as OWC COULD BE using the same number of channels as they did in the 34nm drives. Unlike OCZ, which halved the channels causing some pretty bad performances.

 

[AndrewLupin]

25nm is a bad thing...isn't it???

 

[Guy Mancuso]

this may be shocking, but I only started posting in MR after buying my 2011 MBP a few weeks prior. as to my posting history on this board, well, it's quite extensive and travels far beyond this idiotic thread.

you're quite funny - apparently pulling the curtain away from the smoke and mirrors that makes OWC SSDs 'better' than similar Sandforce-based SSDs is anathema to this thread? how quaint. perhaps i'll make up for it by purchasing their likewise overpriced rebranded MBP RAM modules that run at slower speeds than competing options.

timyu: from an article on Anandtech (I forget where), IBM has the best reliability for their SSDs by far, in part due to extensive testing. OCZ appears the middle of the pack, but with promises to improve -

http://www.anandtech.com/show/4202/the-intel-ssd-510-review/3

Interesting as i own a forum that runs on V Bulletin just like this we do have a post count under each persons name . Surprised that is not implemented here actually. I'm far from a newbie here myself but only post when I see relevant threads that I am interested in just like anyone else.

BTW that forum is this if anyone is interested www.getdpi.com

 

[RainCityMacFan]

25nm is a bad thing...isn't it???

Unfortunately there is a lower lifespan and performance regarding the switch.

But the lifespan will far exceed that of the drive/computer's life and the performance (not too sure on this) is negligible with a good controller. Thus, drives that use 25nm NAND require more over provisioning.

However, the chips are cheaper. Perhaps this is reflected in OWC's change in price but nonetheless, 34nm NAND haven't been in production for months. Stocks are dwindling and it's becoming increasingly harder for manufacturers of SSDs to find 34nm suppliers, thus most are making the transition to 25nm chips.

I wouldn't worry about the size of the chips per se but the company that sells the SSDs that use them. Whether it be OCZ, OWC, or Intel, just find a company that you has something you value. Be it amazing performance, excellent customer service, or unparalleled reliability.

 

[endless17]

Interesting as i own a forum that runs on V Bulletin just like this we do have a post count under each persons name . Surprised that is not implemented here actually. I'm far from a newbie here myself but only post when I see relevant threads that I am interested in just like anyone else.

BTW that forum is this if anyone is interested www.getdpi.com

https://forums.macrumors.com/search/?searchid=22974838

this bot apparently posts in quite a few threads. regardless, my comments on OWC stands. they may be a fine bunch of people, but i don't what justifies the sizable markups on what they sell compared to other manufacturers/resellers.

 

[Guy Mancuso]

I read the same thing on a Intel 510 review and the 25nm did give me cause for pause. Very Interesting cheaper chip shorter lifespan but here is the rub most of us when upgrading to new machines take our SSD with us and put in the spinner again that came with the system to start. My SSD OWC has been in three MBP's already

 

[RaceTripper]

Interesting as i own a forum that runs on V Bulletin just like this we do have a post count under each persons name . Surprised that is not implemented here actually. I'm far from a newbie here myself but only post when I see relevant threads that I am interested in just like anyone else.

BTW that forum is this if anyone is interested www.getdpi.com

You can find anyone's post count by viewing their public profile.

 

[Weaselboy]

At least OWC is advertising the changes on their website rather than marketing it as 120GB.

But I don't think there will be noticeable performance degradation as OWC COULD BE using the same number of channels as they did in the 34nm drives. Unlike OCZ, which halved the channels causing some pretty bad performances.

Where is this advertised on the OWC website? If I go there to buy a SSD and click through all the various purchase options I don't see anything where they disclose they have switched to 25nm. If it is there, I missed it?

I agree with you they have switched and I saw the Diglloyd article but I don't see it disclosed on the OWC site.

 

[Guy Mancuso]

Where is this advertised on the OWC website? If I go there to buy a SSD and click through all the various purchase options I don't see anything where they disclose they have switched to 25nm. If it is there, I missed it?

I agree with you they have switched and I saw the Diglloyd article but I don't see it disclosed on the OWC site.

Neither do I and looked pretty hard to see any description of that. I can't seem to find anything that says switch to 25nm. I would certainly ask if buying

 

[Guy Mancuso]

You can find anyone's post count by viewing their public profile.

Thanks i should have known that too. LOL

 

[Weaselboy]

I read the same thing on a Intel 510 review and the 25nm did give me cause for pause. Very Interesting cheaper chip shorter lifespan but here is the rub most of us when upgrading to new machines take our SSD with us and put in the spinner again that came with the system to start. My SSD OWC has been in three MBP's already

If you are referring to the Anandtech review, the Intel 510 has 34nm NAND and the 25nm he mentions is what will be coming in the Intel replacement for the X25-M. This gen. 3 replacement I believe will be the Intel 320 SSD.

 

[AndrewLupin]

thanks...

so...can we expect a price drop due to this step from 34 to 25???

or maybe thinking about the new sataIII ssd???may be sataII's prices will be cheaper...

 

[Weaselboy]

You can find anyone's post count by viewing their public profile.

You can also get an idea of post count by the number just below the members avatar. For example mine says 6502a, which means a post count above 500.

From the FAQ.

User titles are based on post counts:

Newbie => 0 - 29 posts
Member => 30 posts
Regular => 100 posts (minimum required to post in the Politics, Religion, Social Issues forum)
6502 => 250 posts (minimum required for access to the Marketplace forum)
6502a => 500 posts (minimum required to have an avatar)
65816 => 1,000 posts
65832 => 1,500 posts
68000 => 1,505 posts
68010 => 2,000 posts
68020 => 2,005 posts
68030 => 2,500 posts
68040 => 3,000 posts
601 => 5,000 posts
603 => 10,000 posts
604 => 15,000 posts
G3 => 20,000 posts
G4 => 25,000 posts
G5 => 30,000 posts

 

[Guy Mancuso]

You can also get an idea of post count by the number just below the members avatar. For example mine says 6502a, which means a post count above 500.

From the FAQ.

Okay we actually do names when people hit a certain post count. Like senior member and such. Okay got it

 

[Guy Mancuso]

thanks...

so...can we expect a price drop due to this step from 34 to 25???

or maybe thinking about the new sataIII ssd???may be sataII's prices will be cheaper...

I think that probably will be the case and lets assume so as the Vertex 3 is guesstimated at 499.00 and we see OWC if correctly went to 25nm on there 2 drives than yes it should be less. Sata II drives will drop like flies though as more computers are able to handle Sata III that is just a natural progression of tech.

I should ad the Intel 510 is over 600.00 comparatively to the Vertex 3 so that is also a clue on pricing with 25nm and 34nm