As an owner of a one-day old X25-M, I'm kinda relieved to find out that the write speed has not been bumped up (meaning I am not missing out on much)But of course, we will need to wait for final specifications to be revealed...
-
Jayayess1190 Waiting on Intel Cannonlake
I want an ssd for my new Acer 3810T that I'm about to recieve. It comes with a 500GB HDD, but I want to replace that with one of these:
80GB Intel 34nm
64GB Runcore Pro IV
64GB OCZ Vertex Turbo
64GB Cru-cial M225
64GB Corsair Extreme
Which do you guys think uses the best overall (price, performance, power usage)? I right now am using a 16GB Mini pcie Runcore SSD in my EEE, so I know they are legit. -
If Intel will go for 240 $ I don't see any reason to think twice.
-
Jayayess1190 Waiting on Intel Cannonlake
Nice, out of them all that is the one I want the most. -
Samsung has 30nm NAND already in production, so it's not like Intel has the corner on the smaller-process NAND market.
-
Do you know when the Samsung SSDs based on this NAND will be out? I'm looking to buy an SSD sometime before this Christmas. If the Samsung ones are significantly bigger and at a lower price, they might be worth it (over the 80 GB Intel).
-
Holy crap at those Intel benchmarks! Intel, release these suckers and take my money!
-
My Vertex benches better than that, I certainly hope that's not right. -
I'd like 2 new 80 giggers in RAID 0
The SATA interface would be the bottleneck in that scenario! -
Summit is using the PB22-J Samsung controller and not 2 x Jmicrons.
-
Should have read APEX, not summit.
-
You're joking, right? The only G2 benchmark, and I'd say it's a bit suspect, is pitiful.
I really hope that just because they're cheaper doesn't mean they're gonna be terrible performers like that benchmark shows.
I'll be optimistic, though. Let's wait for real numbers by sources we trust to see whether or not these things are the shiznik. -
im not surprised at the results given its a preproduction unit. i still have faith.
-
I thought the writes were increased to 90MB/s or something like that...
-
As I posted previously, apparently Intel's own spec sheet for postville indicate that 4KB random writes have doubled from 3,300 IOPS to 6,600 IOPS for the 80GB drive and 8,600 IOPS for the 160GB drive. I believe this is what Mr. Wonderful was referring to.
I shouldn't have even linked to the ATTO benchmark that you are referring to. It obviously isn't accurate. We're talking about Intel here, not OCZ...they aren't going to release their a new product that isn't fully tested and free of glaring bugs. -
SATA-II bus speed is about to be broken soon. Photofast Gmonster-V5 now has 270MB/sec read and write
http://hothardware.com/News/PhotoFast-GMonsterV5-256MB-SSD-Sneak-Peek/
Here's a video with CrystalDiskMark benchmark and disassembly:
http://vodpod.com/watch/1861894-photofast-g-monster-v5-raid-ssd
Pretty good for a JMicron drive. Boo on small writes however. -
Indilinx controllers, not jmicron. You can tell from the pics.
-
ssd-s have indilinx but raid card is on jmicron
-
Look at Intel's SSD page: http://www.intel.com/cd/channel/reseller/asmo-na/eng/products/nand/feature/index.htm . Intel doesn't list the MLC G1s anymore which means that there is no reason to buy a G1 over a G2. If the G2s were inferior to the G1s, Intel would sell them side by side at different price points.
Of course, it is best to wait for a complete review (I've been checking AnandTech as they're generally pretty thorough), but it would be boneheaded of Intel to outright replace a product with an inferior version for a slightly cheaper price -- they've never done it with processors. -
The 90MB/s and 32MB wear levelling buffer was from a site that was taking pre-orders on the product. Considering how original Generation 1 "Ephraim" drives used to be 120MB/s for X25-E and 50MB/s writes for X25-M, it might increase. But again, it might not.
All the other infos I posted are 100% correct though. I like the encryption, and server people will like the power safe write cache. -
davepermen Notebook Nobel Laureate
if they went from 50 -> 80 in gen1, then they would go from 90 -> 144 in gen2
(dave likes numbers..
)
-
Haha. I do too.
Intel is becoming PRETTY GOOD at keeping the "small" details on the future products. At this point of release, I should have known more of this but the details are still pretty scant.
Anticipating highly for Braidwood...
-
Me too - if the X25-M 80GB drops to around $250, I will have very little resistance to purchasing.
-
Althernai,
the Intel web page is still for the older ssds.
I can't wait for this new Intels to come out! -
If the new Intel drives show significant improvements, I'll be off the fence. I'm still curious to see what Sandisk has to offer in terms of price / performance with the upcoming G3 drives, but I'm getting tired of waiting. Haven't seen any news on those since the beginning of the year.
-
I don't think so. For example, for the 80 GB standalone MLC drive, they list SSDSA2MH080G2C1 . This is different from the old, expensive drive on Newegg ( SSDSA2MH080G1), and the same as the new, cheap one on FadFusion.
More generally, the new ones are labeled G2 whereas the old ones are G1. The SLC drives on the website are still G1, but all of the MLC ones are G2. -
Microcenter has already started discounting the gen1 X-25M. $249 in-store or online:
http://www.microcenter.com/single_product_results.phtml?product_id=0309018 -
In store only.
-
Not nearly cheap enough...
-
Ah, but it is much cheaper than Amazon's $314. They're looking to get rid of the first gen drives before more people become aware of the fact that there exists a second gen that is cheaper and better.
-
A few months ago the Intel drive hit its low at $220 USD. Then the NAND price hike raised the prices... while its true that it's cheaper than most places, here in this forum, we are aware that the 2nd gen Intel exists. So I'm just stating that even with lower prices, it's not worth it.
-
Kingston SSDNow V-Series 64gb in the Acer Aspire One 751 netbook
-
Of course, but what I am saying is that they are not going after the people in this forum or those who are equally knowledgeable -- they're trying to sell to people who see a cheap drive and take it without being aware of the second generation.
-
According to this, the new Intels should be out day after tomorrow:
http://www.theinquirer.net/inquirer/news/1433725/intel-ssds-launch-tuesday -
Besides access time, it's about on par w/ a standard HDD.
But expected for a Jmicron.
-
However it's fast enough to be out of stuttering territory, at least in that state of use (Not sure if it's used, new, fresh, degraded, etc). It'll still benefit from access times and much faster random reads.
-
davepermen Notebook Nobel Laureate
i guess this is the perfect ssd to prove that speed is unimportant. thanks to the design of ssds, it's still much faster, even while being, from numbers, just an ordinary hdd.
lets just hope they got the jmicron really under control. if they do, this would get a nice netbook drive (and all old jmicron drives should get the firmware update, and a huge price reduction. finally cheap storage for my home server harrharr
)
-
Jayayess1190 Waiting on Intel Cannonlake
Crucial M225 Prices
64GB $169.99
128GB $329.99
256GB $599.99 -
Do we have any idea what controller these use?
-
Not quite guys...compare those benchmarks from the supposedly "new and improved" JMicron/Toshiba controller found in the Kingston V-series to my OCZ Solid series (old POS JMF602B controller)
The so-called "improved" JMicron controller found in the V-series actually has worse results than my POS OCZ Solid.
I really don't understand why most people use the 3 most useless programs for SSD benchmarking and omit IOMeter
-
jmicrons most likely. but they might be the ones using the updated firmware and not the brand new controller. We can only wait for trusted benchmarks for now.
-
davepermen Notebook Nobel Laureate
Jep. So far people reportet that V-Series has no stutterings. So it is good. Good enough, not "wow". But still better than a hard drive for "light usage".
We'll have to wait and see for someone like anandtech to really test it to it's limits. -
umm... if you're using my results, it is caused by the netbook
my netbook isn't giving out full performance for that SSD.. -
how to tell? I am not sure what am I supposed to be looking at here.
-
If you really want to compared the scores.. i ran the benchmark on my desktop.. heres a new score for the Kingston V-Series.. my netbook was capped
as you can see, theres a huge difference when it's in my desktop.. my netbook is just capped for sure xD -
Compared random (4K) R/W speeds. Typical hard drive reads are <1MB/s, while writes are <2MB/s.
-
So compare the OCZ Solid (original sh1t JMicron JMF602B):
With the "new and improved" V-series:
the point is that they are essentially the same performance: terrible.
2.0MB/s in CDM is awful. Keep in mind that CDM runs test for a whole 6 seconds, tests over a 100MB section of the disk, takes the best result out of 5 runs, and uses a queue depth of ONE.
So with those misleadingly light test parameters, the JMicron and V-series controllers both achieve about ~500 IOPS at 4KB random write. If proper test parameters were used it would be around 1/10 of that. If the v-series isn't stuttering, it is just barely getting by under light loads.
Compare those IOPS numbers with the first generation Intel drives which score 3,300 IOPS at 4KB random write (using proper test parameters in IOMeter). In other words, over 300X the performance. -
davepermen Notebook Nobel Laureate
well, those are just numbers. if they 'evened it out', making the quartersec worst writes impossible, then it's still good enough depending on the case.
iops are, as well, not everything.
we're talking here about an ssd which may be the minimum required to have a good usage experience ( not a great one). and this is still nice.
as you said yourself, numbers are not everything. -
meh.. i don't see any stutter in my ssd usage so far xD
-
IMHO these prices are still disappointing... The cost per gigabyte is still too high!
The new SSD Thread (Benchmarks, Brands, News and Advice)
Discussion in 'Hardware Components and Aftermarket Upgrades' started by Les, Jan 14, 2008.