But Intel's pretty ingenious. First off, they're smart enough to know that SF has issues, so maybe they can fix them. Secondly, they're smart enough to have a reason for using SF controllers in the first place. They'll lose customers by the boatload if their SSDs' reliability suddenly plummets.
Mr. Mysterious
-
Mr_Mysterious Like...duuuuuude
-
Yeah I agree. I don`t think we will see SF based Intel drive if Intel is not shure about how to keep the drive reliable
-
-
-
-
-
http://www.pcpro.co.uk/news/369934/25nm-ssd-will-lead-the-industry-says-intel
-
tilleroftheearth Wisdom listens quietly...
Yeah, as far as I'm concerned they're already leading the industry - this will just be the equivalent of getting a bigger HDD in previous years for me.
With the new tech in the latest/most current HDD's, I always got more performance, more capacity and sometimes even at a lower price while also using the same or less power too.
This is where Intel's new SSD will fit in, as an 'evolutionary change' not a revolutionary change (for me).
It may not even have enough real world increase in storage subsystem performance to warrant replacing/upgrading my 250GB 510's just yet...
But I'm hoping it does! -
Tsunade_Hime such bacon. wow
But why would Intel care about the fastest SSD on the market? Intel has been famed for their supreme reliability. Companies like OCZ care about fastest and their SSDs are the first ones to go kaput..
-
-
tilleroftheearth Wisdom listens quietly...
I think the answer for that question is more along like:
Intel has the resources and the capabilities to not only make the most reliable SSD (done! since 2008...), but also the one that excels in any workflow scenario and/or benchmark discipline.
However, the thin line that Intel is walking is one which is balancing that brute force (otherwise causing to be accused/penalized of monopolistic actions) and giving a product that is actually valuable to the consumers and the enterprise (and thereby having adequate return/sales on it's efforts).
It's not a question of either/or (reliability/performance) - it's a matter of balancing and taking everything into account an international organization with the size and clout of Intel to achieve the best possible outcome for all concerned (and especially for Intel), in the end.
This is why I want a 'next gen' SSD from anyone - it will allow/push Intel further along the performance envelope while still keeping the reliabilty crown - and in most real world usage, Intel will still be the performance (overall) champion. -
-
-
tilleroftheearth Wisdom listens quietly...
Samsung. Reliable? Sure. Performance? Not in my experience.
Did I forget them? I'm trying - but you keep bringing them up. -
-
Well, I don't see any reason to forget Samsung, they have decent performances and best reliability so far and with a good price. -
Wonder if this will happen...
-
tilleroftheearth Wisdom listens quietly...
Unless Intel surprises us with a Friday launch of the 520 Series, I would say this is simply more speculation.
How can 'the second quarter in 2011' be pin-pointed to Nov 4th from an unknown/unverified/unofficial (source) slide?
Either way, it looks like an interesting product.
Can't wait (whenever it comes)! -
My God. Can you just shut it with the commenting that everything I post is speculation tiller? I am tired of hearing the obvious.
We all know that Intel may or may not introduce Sandforce allthough something is pointing towards it like mentioned in this thread. And they didn`t just get the 4th november out of the slide of course. Where I don`t know, but they can`t just put out an exact date out of the air can they?
/rant -
-
Dufus, your link is broken.
Yeah, all was pointing towards Sandforce Intel SSD. Another site that confirms it.
I think a teamwork between LSI/Sandforce and Intel can prove to be very satisfying. Intel that knows every single detail with the current and future chipsets, LSI with the enterprice and reliability experience, and SandForce technology that time after time have proven to be one of the fastest, especially benchmark wise. SandForce just recently identified what caused the freezing and BSODs too and fixed it with a firmware. A recent interview with Intel about the Cherryville was also very interesting because Intel said themselves that the Intel 520 will be the fastest one on the market. And we all know how fast the Kingston HyperX is
I welcome this and are looking forward to it.
Btw, in the link you gave Dufus they say that the 520 will be realeased when the X79 chipset is released. So the november 4th is probably wrong then
Thanks for the link -
You can buy X79 now although AFAIK the official launch is the 14th.
Thanks for pointing out the broken link. -
tilleroftheearth Wisdom listens quietly...
Lol... I think you need to relax.
We're both reading the same things/articles and coming to opposite conclusions - nothing to get so tense about - we may both be wrong in the end.
If you call that link 'evidence' or that it 'confirms' that Intel's new SSD will have SF tech inside, well... I think that either I or you have a reading comprehension problem.
But, I thought we were having a conversation here? You're seeing everything through rose-coloured SF based glasses and I'm seeing everything clear (Intel) blue.
Seriously, I want to stop responding... but you keep making these grand jumps in logic that I have to counter with something. -
I am putting more faith in a post from a very knowledgeable Clevo/Sager custom PC builder in the post below than websites trying to be first with breaking news continuously spreading false information. One site starts the rumor and then other sites start quoting the rumor, sometimes verbatim as if the rumor were facts.
I am not picking on anyone in this thread. Hopefully I am providing a more realiable source of information.
http://forum.notebookreview.com/sager-clevo/617053-intel-520-series-ssd-clevo-sager.html#post7984309
-
Mythlogic said that 520 won`t have SF controller 2 weeks or so before the article from Sweclockers was published. Worth to mention.
Anyways, I am excited whatever this turns out to be. Intel controller, Marvell, Sandforce.
@tiller: Im seduced by Sandforce. And very excited. Plus was having a bad day yesterday. Plus in the market for something new for my laptop. Yeaaah -
I really hope its not a Sandforce controller, the 520s were my plan update from x25m, was hopping for new intel own ssd controller, sata III. But if LSI n Intel have faith on Sandforce, maybe ill give in, but the Crucial M4 has been nothing but perfect till today... but will see, still need one more.
-
Tsunade_Hime such bacon. wow
Ya for the things I do with my computer, I really don't need anything beyond G1/G2 for now. Plus are you going to notice the difference between 400 read and 250 read for normal use? Probably not. Also G1/G2 are so cheap now, I bought a brand new G1 80 GB for 105 brand new off of Ebay.
-
Yeah me too Abula. I also plan on upgrading my X25M 160GB.
-
SoundOf1HandClapping Was once a Forge
If I can get something that matches of exceeds my current G2, but with a larger GB count and maybe cheaper, I'll definitely get it.
And to think, this G2 cost me $400 two and a half years ago. -
tilleroftheearth Wisdom listens quietly...
Cloudfire, I'm seduced by technology. But, I've had my rose-coloured glasses kicked in by reality one too many times to take anything in the 'rumor-mill' with less than a few grains of salt.
I too don't really care what controller the 520 will have. It will be subjected to the same standards (mine) as all other components that are a possible upgrade path for my systems. -
-
Tsunade_Hime such bacon. wow
Yeah pretty strange on Ebay old school PM800/810 are still ridiculously overpriced when I saw 2 weeks ago a brand new 256 GB 470 series for 280 shipped..
-
When auctioned, that drive goes as much as $225 from what I've seen. Follow a few auctions and you'll see what I'm talking about. -
Well, the 4th came and went without any Intel announcement.
-
A video of Cherryville SSD vs HDD (I think)
IDF 2011: Cherryville SSD vs. Hardwareluxx | Notebooks Tests, Tipps und News rund um Laptops auf notebookjournal.de - Video Portal powered by Hardwareclips.com -
I'm really getting sick of waiting... Intel shows Q3 2011 on their SSD road map with rumors of a November 4th 2011 release but it's come and gone and still no official announcement, more like Q4 2011.
I'm waiting on Hawley Creek which should be released at the same time or a week after 520 but still no words.
I'm getting tired of waiting for Intel, only reason why I'm doing so is because most on this forum recommend Intel for SSD's especially on mSATA's since Crucial has nothing on the market for this form factor.
I'm so thinking of getting a MyDigital Bullet Proof 128GB mSATA right now, in fact I've been thinking of doing that since a month ago but I keep telling myself it'll be worth the wait for Intel and at the same time thinking if I get the MyDigital and Intel comes out a week or two later that would be bummer.
But this, this is getting ridiculous! -
-
Thanks for comfort zone.
I'm hoping Hawley Creek turns out as good as the 320's speed with inclusion of all features found on it's older brother and without firmware glitch. -
I too am waiting for a worthy SSD to upgrade from my X25-m G2 80GB. From the roadmap articles I've seen, the Intel 520 has 40K IOPS random read, whereas my G2 has 35K IOPS random read. Not much of an improvement.
All the sandforce 2K series SSDs from the comparison charts on Anandtech, the random read performance on SATA 2 isn't that much improved either.
Maybe King Crest will be it, or Sandforce 3K series controllers, or whatever it'll be called -
tilleroftheearth Wisdom listens quietly...
Given any current SSD (since early 2011), the R r/w speeds do not make a drive faster or slower in real world use. Sure the 'scores' on benchmarks change - but that is not how a workstation/notebook computer is normally used.
The response latency of the nand during read/write operations is what is mostly reponsible for the 'feel' of the SSD in actual use.
To see a noticeable difference in how SSD's operate - you need a current platform that can not only show them off, but also not hinder them too.
A SNB platform with an i7 and at least 8GB RAM (16GB RAM recommended) is what will show off SSD differences if there are any.
To put this in perspective:
An Intel 510 Series 250GB SSD (partitioned to use only 100GB and the rest left 'unallocated') on an i7 SNB platform with 32GB RAM 'scores' a 7.8 in WEI. With a 4.4GHz O/C, this system is simply the fastest and snappiest system I have ever used.
An M4 256GB SSD (partitioned as the Intel 510, above) on a mobile i7 platform with 16GB RAM 'scores' a 7.9 in WEI. It is maybe 2-5% 'snappier' than the Intel 510 (with a R w score of ~4.5 times less than the M4 and a R r score ~33% less than the M4) when the storage sub-system is the constraining factor.
When the Intel based platform was also at 16GB RAM - the 'better' M4's storage sub-system R r/w 'scores' were not enough to produce more work - the horsepower of the Intel 510 based better cpu was more than enough to offset the better R r/w benchmarks the M4 could produce.
When the M4 and the Intel were switched and the M4 was used on the higher end platform; there was less work produced than with the nominally lower performing (in benchmarks) Intel SSD - the Sequentials really do play a part in overall performance and it really shows.
When pushed, the current offerings are at least double (in my case closer to 10x) more 'worthy' than an 80GB X25-M - especially if it is a 'gen 1' model - given that the platform and usage we put them on will be both current and 'real world'.
AnandTech - The Crucial m4 SSD Update: Faster with FW0009
So, on a SATA2 platform (and all that entails...) it is not the SSD's that have not improved - it is the platform that is holding them back.
The 'need' and the 'wow' factor of an SSD is best shown on a fully optimized current platform with at least 16GB RAM - and even then, over a state of the art mechanical drive setup, it needs a certain usage scenario to be 'worth it', imo. Unless all we're doing is trying to impress ourselves and our friends with the 'snappiness' our system has.
Point: an i7 SNB platform is simply starving for RAM at 8GB.
16GB makes it come alive (even with an SSD + 8GB RAM initally) and 32GB sets it free. I'm talking just 'snappiness' here - if your workload also demands the high capacity RAM for the smoothest, most lag-free workflow (ie: gaming, video/photo editing, VM's, etc) - then sticking to 8GB (or, horror! less) is like racing at Daytona on Wal-Mart tires.
With prices much more reasonable for 8GB RAM modules, these 'ideal' configs are within almost anyone's reach.
See:
Newegg.ca - Patriot Signature 8GB 204-Pin DDR3 SO-DIMM DDR3 1333 (PC3 10600) Laptop Memory Model PSD38G13332S
See:
Newegg.ca - Patriot Gamer 2 Series 8GB 240-Pin DDR3 SDRAM DDR3 1333 (PC3 10666) Desktop Memory Model PG38G1333EL
Why am I concentrating on everything other than the SSD's in this post? Because the platform+cpu+ram combo make a bigger difference in real world performance (and 'snap') than any current modern SSD does.
When storage sub-systems are RAM (speed-wise) based, I'll change my opinion - depending on the specific workload in question. -
-
tilleroftheearth Wisdom listens quietly...
Dufus,
basically, with WEI (or any other benchmark) what you see (scores) is not what you get (real world).
Even if there are measurable differences in the benchmarks - doesn't mean that real world scenarios take advantage of them. -
-
As soon as you use WEI as a benchmark, I stop reading
. It's a pointless test to run...use a real benchmark program that doesn't give you arbitrary numbers based on a closed Microsoft model that not only outputs meaningless numbers but whose output isn't even correlated to what the user actually experiences.
-
tilleroftheearth Wisdom listens quietly...
Greg, I don't use/rely on any benchmarks except how much work gets done at the end of the day.
WEI may be at the bottom of the benchmark pile - but all benchmarks essentially belong there (at the bottom) for the amount of useful information they allow you to gleam and to extrapolate to real world usage scenarios.
WEI was not the point of my post, in any case.
The overall point was to ickibar123 in post #89 that R r/w 4K 'scores' do not mean a thing to real world performance, given we're on a current SSD and optimally balanced platform. -
I did a long stint in mainframe opsys performance & tuning, and one of the guys I worked with had a saying that sums it up:
"All performance tuning does is change what you wait on."
Meaning: if you have unlimited speed, your job(s) would be finished, not still running. So, tweaking systems always comes down to cpu, I/O(disk) and storage(ram+paging). Crank one component up and now you're waiting on the others to catch up. Usually, I/O is the slowest component and the bottleneck. We're now at an unusual point where I/O is fast enough(sometimes) that the other components are starting to affect things, as noted by tiller's post.
I particularly like ramdrives, as they're a scary fast way to do I/O. But, they do have their particulars that you have to be aware of before using. With 32G ram now commonly available, that's next on the shopping list for my next laptop refresh.
Not a bad spot to be in, ehNever been better, in terms of performance and pricing, for home gamers...
-
Perhaps this was already posted but this seems like an interesting product sheet:
http://www.bouweenpc.nl/wp-content/uploads/2011/11/IntelSSD_520_Series_Product_Brief_Preview.pdf -
Thanks Phil. Haven`t seen that one. Doesn`t say anything about which controller it will be using ARGH! Very nice speed indeed and faster than Samsung 830.
Looking forward to a review of it -
Intel 520 240GB 550/520 MB/sec, 85.000 IOPS max. random write.
OCZ Vertex 3 240GB 550/520 MB/sec, 85.000 IOPS max. random write.
Wanna make any guesses Cloudfire? -
Oh LOL I knew they where similar, but identical? haha gotcha
Intel`s next SSD, 520 - Includes Sandforce
Discussion in 'Hardware Components and Aftermarket Upgrades' started by Cloudfire, Oct 21, 2011.