View Full Version : Bodies / Media E-1 CF Card Read/Write speeds...

02-16-2006, 02:08 AM
Please include your CF card Read/Write results here. At some point they will be put into the chart.
for guidelines on how to do this please see the announcement above.

First Light
02-16-2006, 11:29 AM

Here are the test results for my E-1 with firmware version 1.4:
Lexar 4GB 80x Pro CF card (s/n: 39134GBAI0305A4B6): 3.04 MB/s
SanDisk 2GB Ultra II CF card (s/n: BE0408YMB): 3.28 MB/s

I started testing my cards lasy year with firmware 1.3. Here are the results from the older firmware:
Lexar 4GB 80x Pro CF card (same s/n as above): 2.92 MB/s
SanDisk 2GB Ultra II CF card (same s/n as above): 3.17 MB/s

It has been shown that some manufacturing runs of some memory cards are faster than others. That's why I noted the serial numbers of my cards. If you have an identical card and it measures a little faster or slower, this may be the reason why.

Some size cards are a little faster than others. For example, a 2GB card may be a bit faster than a 4GB card or visa versa. Plus, they use different formatting (cards up to and including 2GB use a 16-bit format; cards over 2GB use a 32-bit format). Therefore, if you want to compare one brand or model to another, it is best to compare equal size cards.

The speeds I measured with my E-1 do not reflect the maximum speed of my two cards because both are faster than the camera (remember, the E-1 is a few years old now). In some of the newer cameras, the Lexar 80x cards are significantly faster than the Ultra II, which (if I remember correctly) is a 60-66x card. This also relates to whether or not a camera fully supports Lexar's WA (write acceleration) technology.

Finally, I ran my speed tests six times each and averaged the results.

02-16-2006, 09:31 PM
E-1 Firmware 1.4

Sandisk Ultra II (s/n BB0406ZQA) =3.200 MByte/s

02-16-2006, 11:27 PM
I am not sure if I have my math right... a little help here please:
I averaged 38.134 seconds for 12 exposures.

Hi Tony,

In your case, the write speed becomes:

(12 x 10.17 MByte) / 38.13 s = 3.200 MByte/s

which I would round off to 3.2 MByte/s because of the uncertainty involved with this experiment.

I'm sorry if I confused everybody by my post about the size of a MByte, that was certainly not my intention.
I just wanted to point out that, for example, an E-1 RAW file is 10.17 MBytes - not 10.41 MBytes - as many intuitively would think when seing the file size 10 412 kBytes in Windows Explorer.

The best is if everybody reports the write speed and the number of RAW files and then one and the same person computes the write speeds. I talk of experience from my years in the MyOlympus Yahoo group where I organized this kind of write speed surveys.

Cheers, Jens.

02-16-2006, 11:37 PM
That's what I thought but then I tried your math... I was confused.
Thanks for helping me out.

02-21-2006, 01:54 AM
(on an E-1 w/ Firmware 1.4)
Lexar 40x 512MB (Gold type or Professional Cards) (s/n:38475120N4603DE4E)
Average time: 41.16 - Average MB/s: 2.970<o></o>
Lexar 40x 1GB (Silver type or Platinum Cards) (s/n: 3784106A84604F9A8)
Average time: 36.58 - Average MB/s: 3.335
SanDisk Ultra II 1GB (s/n: BB05042PB)
Average time: 35.59 - Average MB/s: 3.428<o></o>
RiData 52x / Pro 1GB (s/n: S40002627816)
Average time: 37.37 - Average MB/s: 3.265
RiData Pro-2 80x 1GB (s/n: S5101055962201 8A101G2110104)
Average time: 47.91 - Average MB/s: 2.546<o></o>

I tested each card 6 times like First Light, with formatting between each test.
Interesting the Ridata 80x!! cards are slower the the Lexar 40x….. there’s that WA tech working for yah. Other interesting thing is according to the display on the E-1 the Ridata 80x cards get 97 raw images…. while all the other 1GB cards get 96…:hmm:

02-21-2006, 10:00 PM
San Disk Extreme III 4.0 GB (BH5112ZB) = 4.712 MB/s
I did it 6 times I couldn't believe its speed.

(card DB updated to this point)

First Light
02-22-2006, 12:26 AM
Wow Tony!!! That's fantastic! I put off buying an Extreme III card last year because no one at DPReview thought they would make a difference in an E-1. People were posting amazing speeds in the E-300 but no one had actually tested one in an E-1.

Then I heard a couple of months ago that it did improve the write speed of an E-1, but no one expected such a big difference. My 4GB Extreme III card should arrive in a few days and I'll see if I can duplicate your results. I sure hope so!!!

02-22-2006, 01:22 AM
Well, my E-1 does have new guts too. When they did the shutter, I think they just about replaced everything on it. I was getting horrible write speeds. After I got it back it was cooking along like when I first got it. So that may help too. :dontknow:

First Light
02-23-2006, 09:44 PM

My SandDisk 4GB Extreme III CF card arrived from B&H today and I tested it. I'm "extremely" pleased to report similar fast results as Tony. I've added my results to my earlier test results below. All of these write speed tests were done in my E-1 with firmware version 1.4 and were repeated six times and averaged.

Lexar 4GB 80x Pro CF card (s/n: 39134GBAI0305A4B6): 3.04 MB/s
SanDisk 2GB Ultra II CF card (s/n: BE0408YMB): 3.28 MB/s
SanDisk 4GB Extreme III CF card (s/n: BH05112ZB): 4.89 MB/s

I have to admit that I'm shocked to see that the Extreme III card is so much faster. It is 61% faster than my Lexar 4GB 80x card and 49% faster than my SanDisk 2GB Ultra II card in my E-1.

Most of us E-1 shooters had assumed that the E-1 couldn't write to a CF card any faster than about 3.3 MB/s since both the Lexar 80x and SanDisk Ultra II cards (both capable of much faster write performance) maxed out at this level. Therefore, it seemed logical to conclude that we had reached the limit of the E-1, itself.

Not so!!! For whatever reason (maybe the parallel processing used) the Extreme III card is a lot faster and breaths new life into our aging pro body. This is great news for those of us who shoot mostly raw files and frequently fill our buffers with continuous shooting.

02-23-2006, 10:15 PM
Wow! so mine isn't a fluke. I was going to retest, but you got the same thing! Great news.

04-13-2006, 09:37 PM
SanDisk 512 mb = 63.4 sec = 1.92 mb/s
SanDisk Extreme 1gb = 37.6 sec = 3.24 mb/s
SanDisk Extreme III 2 gb = 27.0 sec = 4.52 mb/s

05-02-2006, 07:07 AM
Compactflash 1gb (7dayshop.com)
C1 01G - T0 03 - TO (1306)

On E-1 (1.4v firmware): 2.86mb/s

Not great, right? But these cards are selling for £15 at the moment! Bargain!

05-15-2006, 01:01 PM
Folks, I'm going to do my cards in a while but thought I'd mention that for any validity the iso should be set to a standard value since the file size changes with a change of iso. How'bout iso 200 just in case there are bodies that don't do iso 100 .... are there?

05-16-2006, 01:17 AM
:nerd: I did mine at ISO 100, and I don't know of any Olympus cameras that don't do ISO 100.

05-16-2006, 06:32 PM
iso 100 it is then!

05-18-2006, 07:05 PM
Folks. How is everyone doing the math? Are you all using 10.17 for file size?

So am I following this right? Push the trigger and the stopwatch at the same time, stop the stop watch when the light goes out. Take the file size and multiply it by 12 and then divide that number by the seconds?

Is that how all of you are doing it - or are you all pushing the stopwatch when the light goes on a half second or so later? Tough to do ... I can't seem to get the coordination sometimes. :)

PS. Both mac osx and photomechanic are reporting my orf files to be 10.2 MB (10,664,102 bytes)

05-18-2006, 07:59 PM
Here are the results I've gotten with my cards:

Transcend 2 GB (120x) - 27.50 sec - 4.43
Sandisk UltraII 1 GB - 36.14 - 3.37
Sandisk UltraII 2 GB - 36.26 - 3.36
Corsair 2GB (Blue, yellow, purple) - 36.37 - 3.35
PQI 2GB ( 100x) - 37.58 - 3.19
Transcend 1GB (orange color, I think rated at 40X) 42.96 sec - 2.82 (by my math)

What is interesting to me that on my D2h which I tested as well the order was quite different and there were far larger gaps. The transcend 2 gb (120x) was by far the fastest. The PQI was the slowest (bummer I have several of them - they were unbelievably cheap at $42 dollars). The two Ultras were quite different in speed with the 2 GB quite a bit faster than the 1GB whereas with the E-1 they were pretty much identical.

So it's very much a case of mating cards to the body. I haven't done it yet but I suspect that the 1D will have a completely diffent mix.

One other surprise and it's a very nice one for me and perhaps for our future: The D2h, which I've been shooting in uncompressed raw mode all this time, simply because I never thought to change that setting was almost twice the speed in compressed raw!! The file size went from 6.1 to an astonishing 2.6 megs. I guess the color black is easy to squeeze. :)

That leaves me with hope that we may get compressed Orfs soon and that perhaps we may get similar speed boosts.

I also now know that I should change from the UltraII's that I've been using as the first cards to the transcends as they are a bit faster. Great to do some experimentation, eh? :)

First Light
05-20-2006, 08:40 PM
So am I following this right? Push the trigger and the stopwatch at the same time, stop the stop watch when the light goes out. ...
Hi Pavel,

No, the start time is wrong. You start the timing when the read/write light goes on---not when you push the shutter button. There is a very slight delay before the read/write indicator turns on. See the instructions (http://www.fourthirdsphoto.com/vbb/showthread.php?t=1167) at the beginning of the forum.

Note: Doing it the way you did (starting the timing when you press the shutter button) won't make a big difference when testing 12 continuous raw shots because the slight delay between the shutter press and the indicator turn-on is very small compared to the total time of the raw shots. Therefore it will not skew the results significantly. But it will probably affect the results if you were testing the speed of a single jpeg image. IMHO it is better practice to get into the habit of doing it right.

05-30-2006, 02:51 PM
Thanks FL. I had re-read the instructions and caught that between the post and the testing. I forgot to edit my post though. You are right that the delay is insignificant as the time is very short - but I did do it the right way.

The speed is not too bad as a per second transfer. If they would give us a compressed orf option it would be really nice. Firmware? I wonder if that would be possible after the fact now?

07-05-2006, 02:24 PM
I just received my new Ridata 8GB PRO 150x CF card (s/n X610130142E301). Here are my test results for CF write times in my E-1 (firmware 1.4):

Average write time = 29.26 sec (12 raw)
Data write rate = 4.17 MB/sec

Test was conducted as described in the instructions in this section.

Though not as fast as the Sandisk Extreme III I was still very, very satisfied to see these results for an 8GB card which I paid $155 (after $20 rebate)!!!

I will test in my E-300 soon and report those results as well.

07-06-2006, 01:15 AM
That's interesting to hear, as every Ridata card I've tried to date has been as slow as a wet week. I can't get one of those cards at the price you did (they're around US$70 more expensive), but they are significantly cheaper than Lexar or Sandisk.



07-08-2006, 08:54 PM

I was as skeptical as you were based on what I'd seen about the Ridata cards. In this case I just couldn't pass up the deal based on $/GB. The performance was hoped for but I wasn't sure it would come through in the E1.

Getting this price and this performance is more than I ever imagined - obviously I'm quite pleased. Even at $70 (US) more I would be considering this a good choice compared to the Lexar and Sandisk models with equal or better performance.

Hope you find something as suitable for your needs as I did.

01-25-2007, 11:16 PM
oki some new cards to add too my collection

Transcend 120x 2GB (m/n:TS2GCF120)
Average time: 25.97 - Average MB/s: 4.69

:eek: :D

Ok i know this will seem dated in a year when you look at this thread but, I couldn't pass this deal up... This 2gb card cost 30$ on newegg.com
I got 4 of them, if these hold up I'll probably get 2 more to fill up my media wallet.
<table x:str="" style="border-collapse: collapse; width: 169px; height: 48px;" border="0" cellpadding="0" cellspacing="0"><col style="width: 39pt;" width="52"><col> <tbody></tbody><col style="width: 46pt;" width="61"><col style="width: 47pt;" width="63"><col style="width: 44pt;" width="58"><tbody><tr style="height: 12.75pt;" height="17"></tr></tbody></table>