Checking speed on SD cards

J

Joey (f)

New member
OK guys I have a question.
How can you check the speed of the SD card? Are there programs for the T3 or or maybe for the PC that will do this?
I have a Sandisk UltraII 512 which I though it was fast, but since the SD cards are dropping in price I was thinking of getting a 1GB card. I guess I also need a standard to compare against.

thank you
mas
 
VFS Mark at http://www.freewarepalm.com/utilities/vfsmark.shtml can be used on the Palm to benchmark SD card speeds. I know it works with OS 4.x and earlier but I don't know if it works with OS5.x.

There are a number of programs for the PC that measure card speeds when read through a card reader. Although the reader itself is a contaminating factor, as long as you used all cards in the same reader, you'd have relative results that would show which card was faster.

Do a Google search on SD card benchmarks and you'll also find a number of tests that have already been done that might give you enough information without you having to do the testing yourself.
 
It works with OS 5.

My PNY (Toshiba) 256MB SD card in a TE:

File Create: 36%
File Delete: 17%
File Write: 9% (blaaaa....)
File Read: 368%
File Seek: 380%
DB Export: 26%
DB Import: 451%
Record Access: 517%
Resource Access: 468%

VFSMark: 252


My Lexar (Panasonic) 64MB SD card in the TE:

File Create: 227%
File Delete: 124%
File Write: 55%
File Read: 400%
File Seek 368%
DB Export: 173%
DB Import: 423%
Record Access: 464%
Resource Access: 426%

VFSMark: 295


My Lexar (SanDisk) 128MB SD card in the TE:

File Create: 48%
File Delete: 40%
File Write: 31%
File Read: 190%
File Seek: 251%
DB Export: 25%
DB Import: 253%
Record Access: 280%
Resource Access: 259%

VFSMark: 153


(If I remember right, these percentages are based on a Palm m500, but I don't remember what card they used.)
 
Thank you both.
I will test the sandisk ultra II and the regular Sandisk card.
Want t see if there is a difference.
And to know what to buy in the future. It seems that most places they sell the cards do not tell you the real speed.
thank you again.

mas
 
Results on some "other" cards...

I have some other cards which I regularly use (tungsten t3, m515, digital cameras), and I downloaded VFSmark and ran it on 4 of them. I have a Lexar 256Mb (32x speed) card on order, but it isn't here to check right now - I expect that by the time it gets here, someone else will have posted the info.

64MB Sandisk SD card (not ultra/ultraII):

File Create: 286%
File Delete: 189%
File Write: 95%
File Read: 478%
File Seek: 386%
DB Export: 135%
DB Import: 348%
Record Access: 430%
Resource Access: 384%
VFSMark: 301

512 MB Sandisk Ultra II SD Card

File Create: 307%
File Delete: 309%
File Write: 157%
File Read: 736%
File Seek: 1180%
DB Export: 189%
DB Import: 908%
Record Access: 841%
Resource Access: 834%
VFSMark: 606

16 MB Panasonic SD Card (came with a digital camera)

File Create: 292%
File Delete: 160%
File Write: 68%
File Read: 644%
File Seek: 786%
DB Export: 187%
DB Import: 657%
Record Access: 738%
Resource Access: 650%
VFSMark: 464


(Just as a comparison {joke?} I took the oldest card I have here and tested it as well...)

4 MB "JVC" (Toshiba) MMC Card (came with a dv camcorder)

File Create: 169%
File Delete: 172%
File Write: 76%
File Read: 68%
File Seek: 256%
DB Export: 46%
DB Import: 211%
Record Access: 159%
Resource Access: 154%
VFSMark: 145

Hope this helps to add a little info - in use the 512 MB Sandisk card does seem a lot quicker than the 64 ... guess I see why now ...

Hope this helps,
 
Merlin, which app do you use to get you benchmark?
I tried using vsfmark, but it seems not to run. I reinstall it a couple of times but nothing. It tells me to wait a couple of minutes but in the end nothing seems to be happening.


mas
 
Last night I went to the website and downloaded VFSmark 1.0 - and it ran without a hitch (with or without Lightspeed activated.

It did *not* want to run when I had TealMaster running with the HBX hack active, however - and I *think* that it didn't like memhack either (but I'll have to check that tonight).

In case it makes a difference (I haven't looked into conflicts yet) I also have the beta of MsMount running on my T3, and I also use PowerRun... Doubt if it makes a difference, but I haven't checked with __ALL__ of the above disabled, so I include the info.

Good luck - I hope my info helps.

(BTW, the VFSMarks do not change when I toggle Lightspeed on/off.)
 
thanks merlin you gave me an idea.
OK I got it to work, Zlauncher was giving me dificulty. I exit the laucher and it worked
T3
results are :
Sandisk 256 mb

File Create: 50%
File Delete: 24%
File Write: 28%
File Read: 317%
File Seek: 472%
DB Export: 27%
DB Import: 393%
Record Access: 425%
Resource Access: 384%
VFSMark: 235

Sandisk Ultra II 512

File Create: 857%
File Delete: 531%
File Write: 89%
File Read: 736%
File Seek: 1072%
DB Export: 341%
DB Import: 930%
Record Access: 841%
Resource Access: 834%
VFSMark: 692
this one seems very close to your mark merlin

I wonder how other 512 cards hold up?
mas
 
SanDisk 512 in a T3:

File Create: 330%
File Delete: 148%
File Write: 55%
File Read: 610%
File Seek: 983%
DB Export: 48%
DB Import: 803%
Record Access: 603%
Resource Access: 600%

VFSMark: 464
 
mspadaro wrote:
> I wonder how other 512 cards hold up?

Funny coincidence - I was wondering the same thing last night so I wrote a quick program to google the web to find the info and then reformat it into a spreadsheet.

The program came up with 53 "hits" and after a little formatting I put them into a spreadsheet which I put up on a web page.

The data is, of course, suspect and has not been verified - it is just what various people have said they have measured.

The URL is
http://www.speakeasy.org/~waustin/SD-MMC-COMPARISON.html
and take it with a grain of salt...

Please remember - this is for entertainment only - don't believe any of the numbers unless YOU have PERSONALLY run VFSMark on the given card in YOUR device :-)

- Merlin
 
Wow!

(Sorry to add after my own post, but this was a surprise to me.)

After I put up the web page, people at my office started running the benchmark themselves, and a few people emailed me their results, so I'm going to leave it up for now - and it now has a few more entries.

However, the news is that I bought a Lexar 256MB 32x card today, and it's numbers are way up on the high end:

File Create: 1174%
File Delete: 1000%
File Write: 53%
File Read: 946%
File Seek: 1072%
DB Export: 155%
DB Import: 897%
Record Access: 1064%
Resource Access: 1037%
VFSMark: 822

This is the best I've seen (2nd is a Sandisk Ultra II 60x card with a VFSMark of 767 - better than I got on mine which was pretty heavily fragmented).

Anyway, I'm sure glad I bought the card - it is noticably faster when I nead large databases from the card (I use PowerRun).

Hope somebody finds this interesting... or maybe even useful.

- Merlin
 
The results on my DigiFILM 256MB SD card are as follows:

File Create: 472%
File Delete: 306%
File Write: 132%
File Read: 725%
File Seek: 1072%
DB Export: 252%
DB Import: 726%
Record Access: 670%
Resource Access: 629%
VFSMark: 553

The test was run on a T3.
 
Mike, you might also take a look at this site

http://www.speakeasy.org/~waustin/SD-MMC-COMPARISON.html

Currently I have over 50 cards on there (but I did not track costs - just performance).

Looking at the duplicate card entries (your page and mine) it would appear that the VFSMark is influenced by both the machine it is run on AND the degree of fragmentation of the card at the time of the test, the best results being obtained immediately after reformatting the card.

Interestingly (to me, anyway), running Afterburner or Lightspeed on the Palm device does not seem to make a difference (on a 515 or a TT3, anyway). I also have FastCPU, but haven't used it for a while, so I don't know whether it affects the reslts :-/

- Merlin
 
Nice spread sheet you have there, some interesting stats as well.




I'm not sure that is always the case. The device does make a difference, however I don't seem to be getting different stats now (60% full) than I did when I first put the card in.



I wouldn't expect a big increase in card benchmarks from overclocking the CPU. The interface to the card and the available bandwidth is still the same, even if the CPU is running faster. You get similar results when you overclock a PC, it does run faster in benchmarks, howerver if you look at hard drive access times, etc. you don't see a big improvement.

Mike
 
Back
Top