16GB SDHC card .. who is upgrading there 8?

[adenicol]

guess it would be easier to just get the 16 gig from the start. Thanks for the info.

[stroths]

I am contemplating a 16gb class 6, but have a hard time justifying $60 for it. I can get a 16GB class 2 for about $30 right now on OfficeDepot's website, but I don't want a class 2.

I may just use my current 8gb class 4 until the class 6 come down a bit.

[TheBundo]

I'll take the class 6 16 gig card out of my TP2 before I sell it, and put in an old small card. I already have 13 gigs full on the TP2, so I can only imagine a 32 gig will be needed before years end.

[themuffinman]

Well, I just bought 2 16gb class 6 cards from ebay for $50 each with free shipping. Damn I can't for this phone to come out, 6 more days.

[stroths]

I posted this in another thread, but it applies here as well:

I was going to wait for a class 6 16GB Micro SD, but I got such a great deal on a Sandisk 16GB Micro SD (Class 2) that I could not pass it up. I got it from Office Depot for $23.76 out the door.

They have it on their website for about $32, but it's delivery only. They carry the same model in store, but its a different SKU and priced at $99. I asked them to price match their website which they did with no problem, which surprised me since the SKU was different. I will mention that they are the exact same model though (SDSQ-016G-A11M) regardless of the SKU.

Anyways, I used THIS office depot to get $10 off my purchase. It may not be super fast, but it should be very reliable being that it's a Sandisk and it was dirt cheap.

Tested it with Crystal Disk Mark on my PC with a card reader and got about 18mb read and 9mb write on the sequential and 512k runs which isn't too bad for a $23 16GB MicroSD

The 4k weren't as great, but still I can work with this card. Below are the benchmark details.

EDIT: I added benchmarks for my 8gb Kingston Class 4 that I use in my TP2. The benchmarks on the 16gb Class 2 actually look better. That is odd.

EDIT 2: I added some more test results to ensure my results are accurate.

-----------------------------------------------------------------------
CrystalDiskMark 3.0 x64 (C) 2007-2010 hiyohiyo
Crystal Dew World : Crystal Dew World
-----------------------------------------------------------------------
* MB/s = 1,000,000 byte/s [SATA/300 = 300,000,000 byte/s]
OS : Windows 7 [6.1 Build 7600] (x64)

Here is my Sandisk 16GB Class 2 Micro SDHC

Test 1: 50 MB [L: 6.7% (1.0/14.8 GB)] (x1)

Sequential Read : 17.957 MB/s
Sequential Write : 8.971 MB/s
Random Read 512KB : 17.961 MB/s
Random Write 512KB : 2.092 MB/s
Random Read 4KB (QD=1) : 4.004 MB/s [ 977.5 IOPS]
Random Write 4KB (QD=1) : 0.022 MB/s [ 5.3 IOPS]
Random Read 4KB (QD=32) : 4.061 MB/s [ 991.4 IOPS]
Random Write 4KB (QD=32) : 0.033 MB/s [ 8.1 IOPS]
-------------
Test 2: 100 MB [L: 6.6% (1.0/14.8 GB)] (x5)

Sequential Read : 17.914 MB/s
Sequential Write : 9.358 MB/s
Random Read 512KB : 17.922 MB/s
Random Write 512KB : 2.131 MB/s
Random Read 4KB (QD=1) : 4.020 MB/s [ 981.4 IOPS]
Random Write 4KB (QD=1) : 0.020 MB/s [ 4.9 IOPS]
Random Read 4KB (QD=32) : 4.096 MB/s [ 1000.0 IOPS]
Random Write 4KB (QD=32) : 0.027 MB/s [ 6.6 IOPS]
-------------
Test 3: 500 MB [L: 6.6% (1.0/14.8 GB)] (x2)

Sequential Read : 17.921 MB/s
Sequential Write : 9.434 MB/s
Random Read 512KB : 17.816 MB/s
Random Write 512KB : 1.950 MB/s
Random Read 4KB (QD=1) : 3.555 MB/s [ 867.9 IOPS]
Random Write 4KB (QD=1) : 0.019 MB/s [ 4.7 IOPS]
Random Read 4KB (QD=32) : 3.868 MB/s [ 944.3 IOPS]
Random Write 4KB (QD=32) : 0.020 MB/s [ 4.9 IOPS]
---------------------------

Here is my Kingston 8GB Class 4 MicroSDHC (currently used in my TP2)

Test 1: 50 MB [L: 46.5% (3532.5/7592.0 MB)] (x1)

Sequential Read : 18.550 MB/s
Sequential Write : 8.953 MB/s
Random Read 512KB : 18.553 MB/s
Random Write 512KB : 1.620 MB/s
Random Read 4KB (QD=1) : 4.011 MB/s [ 979.2 IOPS]
Random Write 4KB (QD=1) : 0.017 MB/s [ 4.2 IOPS]
Random Read 4KB (QD=32) : 3.901 MB/s [ 952.4 IOPS]
Random Write 4KB (QD=32) : 0.030 MB/s [ 7.3 IOPS]
----------------

Test 2: 100 MB [L: 46.5% (3532.5/7592.0 MB)] (x5)

Sequential Read : 18.505 MB/s
Sequential Write : 9.547 MB/s
Random Read 512KB : 18.479 MB/s
Random Write 512KB : 1.978 MB/s
Random Read 4KB (QD=1) : 3.860 MB/s [ 942.4 IOPS]
Random Write 4KB (QD=1) : 0.017 MB/s [ 4.2 IOPS]
Random Read 4KB (QD=32) : 3.865 MB/s [ 943.6 IOPS]
Random Write 4KB (QD=32) : 0.024 MB/s [ 5.8 IOPS]
--------------

Test 3 : 500 MB [L: 46.5% (3532.5/7592.0 MB)] (x2)

Sequential Read : 18.512 MB/s
Sequential Write : 10.413 MB/s
Random Read 512KB : 18.181 MB/s
Random Write 512KB : 1.724 MB/s
Random Read 4KB (QD=1) : 3.942 MB/s [ 962.3 IOPS]
Random Write 4KB (QD=1) : 0.017 MB/s [ 4.2 IOPS]
Random Read 4KB (QD=32) : 3.891 MB/s [ 949.9 IOPS]
Random Write 4KB (QD=32) : 0.018 MB/s [ 4.4 IOPS]
-------------------
Here is a Kingston 8GB Class 4 regular size SDHC (not micro)

Test 1 : 50 MB [L: 0.0% (0.1/7592.0 MB)] (x1)

Sequential Read : 16.853 MB/s
Sequential Write : 9.397 MB/s
Random Read 512KB : 16.826 MB/s
Random Write 512KB : 1.681 MB/s
Random Read 4KB (QD=1) : 4.074 MB/s [ 994.6 IOPS]
Random Write 4KB (QD=1) : 0.039 MB/s [ 9.6 IOPS]
Random Read 4KB (QD=32) : 4.089 MB/s [ 998.3 IOPS]
Random Write 4KB (QD=32) : 0.063 MB/s [ 15.3 IOPS]

[gTen]

Your tests were only done 1x...you need to do more then that to get accurate results...also you'd want to try different file size samples based on usage.

But yes Class only stands for the "minimum" speed..the maximum speed could be on par with a Class 6...but for high video you want to make sure to have a good minimum to prevent shutter.

Goes to one of my favorite sayings

"Then there is the man who drowned while crossing a stream that was, on average, 6 inches deep"

[stroths]

Thanks for the advice gTen. Just updated my stats and I see a little bit of a speed increase now. I am pretty satisfied with my purchase considering the price.

[Drac]

Originally Posted by stroths

I posted this in another thread, but it applies here as well:

I was going to wait for a class 6 16GB Micro SD, but I got such a great deal on a Sandisk 16GB Micro SD (Class 2) that I could not pass it up. I got it from Office Depot for $23.76 out the door.

They have it on their website for about $32, but it's delivery only. They carry the same model in store, but its a different SKU and priced at $99. I asked them to price match their website which they did with no problem, which surprised me since the SKU was different. I will mention that they are the exact same model though (SDSQ-016G-A11M) regardless of the SKU.

Anyways, I used THIS office depot to get $10 off my purchase. It may not be super fast, but it should be very reliable being that it's a Sandisk and it was dirt cheap.

Tested it with Crystal Disk Mark on my PC with a card reader and got about 18mb read and 9mb write on the sequential and 512k runs which isn't too bad for a $23 16GB MicroSD

just went to an office depot and they wouldn't price match.

[stroths]

Sorry it didn't work out for you, but it's always a gamble with a pricematch like this one. If you don't mind paying around $33 shipped, you can get a similar one here - LINK

------------------------

Another thing I should mention (unrelated to the price of the card) is that cluster size can make a difference in how the card responds. The smaller the cluster size, the less wasted space you have. However, it also slows the responsiveness of the card down overall. A larger cluster size (i.e. 32k) will make your card more responsive, but will also waste more space. For example, a 1k file will take up the full 32k in the latter scenario. If you had 4k clusters, you would only be using 4k for that same file.

If you have a lot of files less than 32k then you might be better off formatting the card with 4k clusters. Otherwise, I would suggest the larger cluster size of 32k to get a better response out of your card.

I have done this in the past with slow SD cards to get a better response. For example, on my hacked GPS unit with all the maps on the SD I used the larger 32k clusters from the default 4k. After doing that and copying my files back over, I had next to no lag. With the 4k clusters, I would regularly get lag as the GPS software was reading the map files.

I have also done this on my media server, but I used 64k clusters there. This was suggested over at the SageTV forums (which is what I run on my media center). This got rid of stuttering I was experiencing on HD movies and tv shows.

To format to a new cluster size, simply right click on the drive in Windows and choose format. Choose quick format, FAT32, pick your cluster size and I would suggest you check the "quick format" unless you have a lot of time to burn. WARNING: A format will wipe all data from the card, so backup anything you need. If you don't have a way of mounting the card on a PC, you might google to see what other options you have for formatting.

If your want to see your current cluster size in WIndows, go to a command prompt and type "Chkdsk X:" (without quotes) where X is your MicroSD drive letter. You will get some outpul that will include something like mine 32,768 bytes in each allocation unit. This means mine is currently at 32k (32,768 bytes).

There's lots of literature you can read online about this and its pretty common. Below is a quote I pulled from one site that explains why larger clusters can increase performance.

What is an issue is how fast your system can read and write files from the hard drive. The larger the cluster size - the larger "chunks" that are fetched from the drive. This speeds up performance. For example, many drives were formatted with 4k clusters. This means to read 32k of data, the system must perform eight fetches - 4k per fetch. If you had instead formatted the same drive using 32k clusters, the system would only have to perform one fetch.

[adenicol]

I hope I didn't just screw myself...bought a 16 gig no-name class 6 card on ebay. I hope It's not a piece of crap... Can someone give me their opinion on this? http://bit.ly/d9T6y6