SSD Thread (Benchmarks, Brands, News, and Advice)

I did. But nothing changed

I gave OCZ a try mainly because of 4k performance. Otherwise I would have picked up a Intel 510.

Do you guys think I should return this and get a 510 instead ??

 

You're getting better 4K performance than the Intel 510 could ever give you.

In other words, no I don't think you should get Intel 510.

Intel 510 is interesting when you do large volumes of file copies of incompressible files. Like Tiller.

 

Not sure if I would buy that argument. Why would laptop be slower for only that particular case whereas the other numbers don't show similar drop ?

Interested to hear the reason.

 

Have you tried to combine my tweak with ThrottleStop in order to disabel EIST and C1E? I was able to get 100% of my SSD's performance, check here
Maybe it will work for you as well...

 

Thank you all for your help I honestly dont want to touch those tweaks because I dont want my lap to be any hotter. I tried just to confirm if the R3 still has the same throttle issue with 4k random like the R2 had.

 

I browsed a little on Ebay looking for SSD's and found a few deals.
I just wanted know anything for/against these SSD's:

Intel X25M160G (160 GB, SATA) Hard Drive | eBay
Intel X25M160G (160 GB, SATA) Hard Drive | eBay
Intel X25M160G (160 GB, SATA) Hard Drive | eBay

Anything i should be aware of? Something that decreases wha they're worth?

 

0 feedback, Member since Apr-10-11, "single second boot times", "only $10.00 international shipping".

I have very little confidence in this seller. I wouldn't buy from him. I don't know much about Ebay's buyer protection.

I read Newegg has a $299 deal on the Intel 320.

 

My thoughts are that the ebay items are hot. If you're cool with that, then go for the risk.

Oh, would I love for business to pick up, so I can pay my taxes, pay my bills, and get the 160GB 320 series.

 

you should buy one and let us know if he's legit.

 

Seems like OCZ is coming out with an extra fast Vertex 3 version dubbed MAX I/O.

"it's not a secret, there is an enthusiast based SKU coming next week we hope. I do not have one yet, i have been told the 120GB version in this range is faster than the regular 120 Vtx3...after that i have no further info."

General Discussion Vertex 3 MAX IO partno

"The difference is that it uses 3Xnm toggle-mode flash vs. 2Xnm sync-mode flash which will yield higher performance. "
http://thessdreview.com/Forums/ssd-discussion/642.htm

So a Vertex 3 with 34nm toggle mode NAND....

 

Hey !

Well, ya, been temporarily deceased lately; slowly coming back to life (springtime )...

This being said, and more importantly: how's your rig doing ? SSD still performing well ?

Mines need to be reformated (quick) soon, 'cuz I'm loosing speed; it's been over 6 months now I'm running them in RAID with obviously no TRIM support, so it's getting slow in there...

On another note, anyone knows if Dellianware is now shippping the 470 series of Samsung's SSDs with their systems ?

Great day to all !

 

good job OCZ. announce that two weeks after the normal ones ship so all the early adopters get screwed.

more bad PR coming their way I bet.

 

Different product at different price point for different purpose, I see nothing wrong with that.

 

I could see some angry customers coming from this. They were expecting to buy the fastest 120GB drive only to find out it was not the fastest a few weeks later.

 

Return/Sell the current v3 then wait for the newest and greatest. The V3 has only began shipping quite recently and very likely to be still within return policy limit.

 

thats exactly my point.

 

This time, when changing kind of NAND flash is a good factor for sells, they are rumoring it very clear

Somebody know when sandforce will deliver its new controller to other companies like corsair ?

 

Good point. That tells you something about the claim that 'oh the support staffs on the forum know nothing but was trying hard to help once they noticed the issue(the 25nm switch)'. classic good cop/bad cop.

BTW, given SF's history I would rather wait for someone to beta test any 'new tech'

Have a feeling that their engineers are firmly in the camp of 'release fast, release often'. That is fine for a browser but not somthing for my primary storage system.

 

That's why i wanted to buy the last pieces of g2 controllers, i didn't want to be a beta tester for Intel or ocz... Sure, c400 and specially vertex3 pwn everything now, but everybody remember freeze problems in some c300 or vertex2 not seen in bios... Most sellers are trying to sell out old SSD but they are still rocket ! Good time to buy them if you stil have a HDD

 

Zero feedback seller with a bunch of high ticket items for sale all at once. Hmm, what could go wrong?

If you're willing to chance it though, as long as you pay by credit card, you can issue a chargeback (you get your money back instantly). I believe eBay also protects your purchase, but may require you to jump a couple of hoops before getting paid.

 

Giveaway of the Day - Paragon Migrate OS to SSD Special Edition (English Version)

It's FREE....

 

Intel and Micron announce new 20nm NAND Flash manufacturing process

 

I came here to post that.

"promises to allow for 128GB of storage in a solid-state storage solution smaller than a postage stamp"

holy crap cant wait.

 

"promises to allow for 128GB of storage in a solid-state storage solution smaller than a postage stamp"

With only 10 writes per cell before it fails.

 

They absolutely have to address the writes issue at this point. Not that they can't guess this for themselves, but simply having more "spare space" isn't going to entirely address the fragility that is becoming inherent in the finer widths.

 

They don't have to guess for themselves, they actually KNOW how much (or not) it matters. unlike the typical enduser, who read about it on the web and how big of an issue it is while no one is affected.

 

The problem with nand cells is that as soon as there's something written on a specific cell, which is part of a particular page, itself part of a said block, any other write to occur on any cell part of the block will be slower than it should because the controler has to read the already written data, delete it, write it in RAM, and then re-write it ALONG with the new data that was to be written...

Therefore, until or unless SSD's manufacturers find a way either to overwrite data on nand cells, or to write to a cell on a page in a block without having to go through the 4 steps described above, then, there will be slowdown coming from writing to partially written blocks. This is because, TRIM or not, there will always be partially written blocks, therefore, some slowdown.

This being said, and while any SSD is WAY faster than any HDD, I can myself feel the difference with my two Intels in RAID (so no TRIM); they slowly become slower, as I write to them. It's not the end of the world, but over a 6 months period say, windows boot time goes from 15 seconds to 25 ~ 30 seconds; launching an app instantly takes now 1 or 2 seconds, you know, not a so big deal, however, I do indeed feel the difference. I' ve been playing with 2 SSDs in RAID for more than a year and a half now, so again never got TRIM on my side, and while my SSDs are still way faster than my previous 7200RPM RAIDed HDDs, I do notice the speed decrease over time.

Fortunately for me, I just have to reinstall windows every six months or so to recover the speed. This is being done by quick formating my drives while installing windows (one of the first screens when you install). While overprovisionning an SSD with more nand cells than advertized (and user-usable) do indeed help, it's not a panacea, so let's hope the technology will evolve and solve this issue.

 

Whad do you mean by block ? erasable block or writable block ?

 

Data is written to cells, which are grouped in pages, which are grouped in blocks. The problem is while the controller can erase an entire block in one shot, it cannot however write to a cell if any other cell within the same block is holding data...

Therefore, the controller has to first delete the already written data, so it can write something onto a cell within the said block. If the controller just deletes that data, you loose it; so, it has to first read it, then delete it, then write it to RAM, and then re-write it along with the data that was to be written.

 

When SSD's get into multi-TB capacities, the write/erase page/blocks will be identically sized - there will be so much nand available that it will be 'affordable' to do that.

Once there, this read/modify/erase/write cycle should disappear once and for all (unless people start filling TB sized SSD's to capacity...).

Just like DSLR camera sensors, more (smaller) photo sites (=more nand capacity) is not necessarily a bad thing (primarily; noise in the final image). Increased photo sites not only offer better resolution and acuity, but they also offer increased colour graduations (colour fidelity) and maybe more importantly increased dynamic range too.

With SSD's, an order of magnitude increase in capacity (compared to today's puny offerings) will mask many more of these physical limits that nand has (glaringly) today.

Not only will it make how many write cycles each nand has less important - it will also enable the use of less sophisticated (read: cheaper) controllers too. In addition, with less 'GC' needed, the promised power saving features of SSD's might finally be (universally) be realized as well.

When SSD's finally get out of their diapers and past their teens - that will be the time when 2TB SSD's are not only affordable - but available too.

Can't wait to meet a (grown up) SSD.

(Hope it happens while I'm still 'young at heart').

 

Um, I am not sure that is the case. I would suggest double check.

 

Source:
Q&A: Tom's Hardware And Kingston On SSD Technology : Kingston Up Close

 

Nice info Phil.

Just pointing out though that this was early 2010 when nand was 50nm and 34nm based.

Not to mention the much lower (Kingston's) 4K R r/w's that greatly affect write endurance. With a strictly random 4K r/w load (the 'harshest' nand can be subjected to); the write limits of all SSD's are nowhere close to that chart.

But it does illustrate nicely how capacity quickly masks all normal (and even my 'un-normal') write needs.

Even though it is 'only' a linear function; it quickly grows past the point of 'consumer' usage patterns and begins to handle professional use quite well.

If manufacturers were clear about the tradeoffs made to get their product to market, we could be making a solid choice of which SSD to purchase for our use.

Intel is the only one who consistently communicates and delivers products that focus on reliabilty just as much as performance. All others are simply chasing the latest 'benchmark' fad to push their stuff (when they're not outright lying).

With all the parameters that can be tweaked, adjusted and manipulated (via the controller/firmware), the raw capacity of SSD's will be a clear indication of the reliablity of SSD's going forward.

 

I learned that before I actually bought my first SSDs, so it's a no brainer for me, it's already checked and double-checked.
I'll however get back to you on this. → See here

 

I'm looking for a 60GB-64GB SSD with TRIM, at least a balanced performer, low power consumption and it must be under £100.

Kingston i've noticed have come out with some new drives. Well, new to me since i last checked lol. I like the V+ but that is just over £100. I'm unsure how the V100 drives are power consumption wise.

Here are some drives i've seen.

Corsair CSSD-V64GB2-BRKT Nova
Corsair CSSD-V64GB2-BRKT Nova: Amazon.co.uk: Computers & Accessories

Integral 2.5 inch 64GB SATA II MLC
Integral 2.5 inch 64GB SATA II MLC Internal Solid State Drive: Amazon.co.uk: Computers & Accessories

Kingston SSD S100 new V series
Kingston SSD V100 64GB SATA2 2.5inch Hard Drive: Amazon.co.uk: Computers & Accessories

OCZ OCZSSD2-1VTX60G
OCZ OCZSSD2-1VTX60G 60 GB Vertex Series SATA II 2.5-Inch SSD Drive: Amazon.co.uk: Computers & Accessories

Which drive do you think fits the bill the most? Integral is unknown brand but it's a cheap drive so i won't overlook it. The Vertex is also very reasonably priced but i'm unsure if it fits all of my criteria.

 

Dave p, what do you think of an Intel 320 160GB for my next drive? My NOVA crapped out badly Thanks.

 

Has anyone seen a roadmop for Intel ssd's that goes past Q1'11?

 

Samsung 470
Samsung MZ-5PA064 64GB 470 Series Solid State Drive 2.5 SSD

Fast, reliable and power efficient.

Not sure about warranty and if that's a good webshop.

 

Thanks Phil
I will look into that. That is a very good price.

 

Samsung 470 could be interesting too.

New review that shows real world performance of the 320 vs. Samsung 470:
OCZ Vertex 3 240GB SSD Review > Benchmarks: Real-World Applications - TechSpot Reviews

 

May be I have misunderstood you so I would try to see if we are talking about the same thing. SSD nowadays more or less are arranged in the following manner :

1M erasable block each contains 4K writable page(if you prefer to call it that).

Each 4K block can be written individually ONCE only. So if a 4K page is deleted(i.e. TRIM) or overwritten(LBA wise), it is no longer writable(even technically still possible depending on previous data but I doubt any controller would bother than). overwritten of a 4K page is done by mapping the LBA to a new NAND location and the original 4K page pointed to by the LBA becomes 'no longer in use'.

Your described read-erase-write back happens only when there is no longer available free(writable) 4K page available. IOW, within a 1M erasable block, there can be lots of 4K page that have once been written but is marked as 'no longer used' and if the SSD cannot find writable 4K page, these 'no longer used' one needs to be reclaimed by erase the whole 1M block and write back those that is still 'in use'. Of course, lots of wear leveling, optimizations(packing, moving around etc.) happen.

 

Interesting. I'm not sure I understand fully thought, but what I understood is this:

Each block of say 1024K contains say 8 pages of 128K;
You can only erase a full block, never only one page inside;
You cannot write on a page containing any (valid or invalid) data;
And this even if the 128K is not fully used, even if there's only 4K written;
Thus, if you need to write would it be only 4K on a page containing any (valid
or invalid) data, the controller has to first erase the whole block before it can write there.
In this case, it would be up to 8 pages, up to 1024K to erase before being able to write the 4K...

This is why manufacturers came up with TRIM, so the data is indeed eradicated when you delete it, where ITGC can only defragment the invalid data...

I don't know if we're on the same page (of 128K ), but this is my understanding.

Cheers !

 

yup. it is the page size that we may read it differently. I am more side with the 4K which is also the natural size used by CPU and the Advance format drive but we may need to get it from the horse mouth. the only thing I can find so far is this;

X25-M(V) erase block size? - AnandTech Forums

 

Great.

So I guess this guy says it all:

The confusion between 128KB and 512KB page sizes probably stemmed because each page is 4KB and there are 128 pages in a block, for 512KB block sizes.


So it would go like this:

Each block of say 512K contains say 128 pages of 4K;
You can only erase a full block, never only one page inside;
You cannot write on a page of a block already containing any data;
Thus, if you need to write would it be only 4K on a block containing even
only one page, only 4K of valid data, the controller has to first erase the whole
block before it can write there. In this case, it could be up to 128 4K's pages, up to
512K to erase (providing there's such invalid data) before being able to write the said 4K...


If I got it correctly, I would like to sincerely thank you for your lights.

If not, well, I dunno, lemme think about it, k ?

 

you can write to any one of the 128 page independently, SO LONG THEY HAVEN'T NOT BEEN WRITTEN TO BEFORE(after an erase). You cannot 'rewrite' it or 'TRIM then write' again. That is the nature of NAND, it is a WRITE ONCE page.

 

OK, then it would go:


Each block of say 512K contains say 128 pages of 4K;
You can only erase a full block, never only one page inside;
You cannot write on a page of a block that has not been erased;
Thus, if you need to write would it be only 4K on a block containing even
only one page, only 4K of valid data, the controller has to first erase the whole
block before it can write there. In this case, it could be up to 128 4K's pages, up to
512K to erase (providing there's such invalid data) before being able to write the said 4K...

 

Each page(4K) has 3 logical state:

1. Writable(i.e. the OS hasn't use it and the block containing it has just gone through an erase)

2. Read Only(i.e. the OS has written to it and the data there is needed)

3. Read Only but 'no longer in use'(i.e. OS has either TRIM it or has rewritten to it which means the new data have been mapped to some other place)

If the page is in (3) and the controller cannot find other pages that it can write to in order to fulfill an OS write request, the controller would need to erase the whole containing block then write back all the pages in state (2) then write the extra data requested by the OS.

 

i have a doubt about NAND scaling and performance:

if they shrink 128 GB as little as a postage stamp, do they will reduce number of channels as OCZ did with vertex2?

today 64 GB SSD are the one with the lowest number of channels used, with this aggressive scaling will they reduce number of channel for larger SSD ( 80, 120, 160 size) ? write performance will be the same more or less ?

I suppose that 20 nm is still reliable for a workstation user, but at this point i would start thinking about lowering SLC price for the future...

 

Well, that is where Intel's muscle may be showing(if they are still interested in this business). There is nothing stop them to package the controller and the NAND together given their mighty process technology. I thought their Ivy Bridge is rumoured to use these stacking technology to embed GPU used memory inside the CPU package.

 

I think we're saying the exact same thing, but just to make sure:​

1. Each page (4K) can be in any of the 3 following state:

→ (1) unused → is empty (unused or erased);
→ (2) used →→ contains valid data;
→ (3) occupied → contains invalid data.

If the controller needs to write to a page in the (1) state, it writes right away no problem;
If the controller needs to write to a page in the (2) state, it won't because there's valid data as per the O/S;
If the controller needs to write to a page in the (3) state, it will need to:

a) Read the data in the (2) and (3) state pages of the block;
b) Write this data to RAM;
c) Erase the full block;
d) RE-Write the previous (2) and (3) state pages along with the data that was to be written at that time.


2. So, it would go like this:

Say each block of 512K contains 128 pages of 4K;
The controller can only erase a full block at one time;
The controller cannot write to a (3) state page without erasing first;

Thus, if the controller needs to write would it be only 4K on a (3) state page within a block, it has to:

a) Read the data in the (2) and (3) state pages of the block;
b) Write this data to RAM;
c) Erase the entire block;
d) RE-Write the previous (2) and (3) state pages along with the data that was to be written at that time.


In this scenario, it could be up to 128 4K's pages, up to 512K of data on
(2) and/or (3) state pages to read-erase-write before being able to write the said 4K...


Now, are we on the same (0) state 4K page ?