dtrayers
14th of November 2003 (Fri), 08:41
It just arrived yesterday and while I was waiting for the laundry to finish I did some speed tests comparing it to a standard PCMCIA CF reader. I ran these tests on an IBM Thinkpad A31 running XP Pro and with 768 MB of RAM. The camera is a 300D.
I have three CF cards, a Sandisk Ultra II 512MB, a Sandisk Ultra 256MB and a Lexar 4x 256MB. I timed the speed of transferring a folder consisting of 16 RAW (.CRW) images and the corresponding .THM files. The folder size was 122,586 Kb (119 MB). I transferred the folder to the CF card, then transferred the folder back to the Windows Desktop. I ran both cards twice with each reader and averaged the times.
As a last test, I shot 22 RAW images (45 for the Ultra II) of the same scene (using a tripod) on each card and tested the download speed using DownloaderPRO. I think this is a little more "real world", as Downloader is adding some overhead in renaming the files. It would add even more overhead if it had to rotate images, which it didn't have in this test.
Here are the results:
http://home.comcast.net/~dtrayers/photos/CFSpeeds.jpg
The Delkin is about twice as fast as the Dazzle in reading the card with the Sandisk, but four times faster with the Lexar. This suprises me, as I thought the Sandisk Ultra was a 'later' technology (at least I purchased it after the Lexar and paid more for it). I'm going to use the Lexar before the plain Ultra, as it clearly is faster.
I was surprised at the speed difference between the write and read speeds of the Lexar. Clearly it performs faster in the read mode.
I compared my results to Rob Galbraith's (http://www.robgalbraith.com/bins/multi_page.asp?cid=6007-6425) CF database and to my surprise they are in agreement with his. The Lexar 4x it quite a bit faster in his Card-to-PC test than the Sandisk Ultra. It seems that the Card-to-Camera speed is limited by the camera. The fasted card he's tested in the Card-to-PC is not a lot faster than the rest in the Card-to-Camera.
Oh yes, and I did one more subjective test while taking the images: I placed the camera into continuous mode and made sure the flash was off and took four successive pictures by holding the shutter release. I measured the time from the press of the shutter to when the CF light went out. For the plain Ultra and the Lexar cards it was an equal 32 seconds. The UltraII was 28 seconds. Since I was shooting the same scene and the file sizes were for the most part the same, I conclude that the cards perform similarly in the camera. This result also agrees wtih Galbraith's, and his tests also include the latest Extreme-Ultra-II-Ludicrous-Speed cards. The fastest card he's tested in Card-to-PC is about 9000 Kb/sec, but it's not *that* much faster than the middle of the pack in the camera. Clearly the 300D is the limiting factor, which Galbraith also concludes.
So from my little test, the Delkin can give some significant speed improvement with the right card. Also, I feel I can use Rob Galbraith's database as a pretty good indicator of card performance.
I have three CF cards, a Sandisk Ultra II 512MB, a Sandisk Ultra 256MB and a Lexar 4x 256MB. I timed the speed of transferring a folder consisting of 16 RAW (.CRW) images and the corresponding .THM files. The folder size was 122,586 Kb (119 MB). I transferred the folder to the CF card, then transferred the folder back to the Windows Desktop. I ran both cards twice with each reader and averaged the times.
As a last test, I shot 22 RAW images (45 for the Ultra II) of the same scene (using a tripod) on each card and tested the download speed using DownloaderPRO. I think this is a little more "real world", as Downloader is adding some overhead in renaming the files. It would add even more overhead if it had to rotate images, which it didn't have in this test.
Here are the results:
http://home.comcast.net/~dtrayers/photos/CFSpeeds.jpg
The Delkin is about twice as fast as the Dazzle in reading the card with the Sandisk, but four times faster with the Lexar. This suprises me, as I thought the Sandisk Ultra was a 'later' technology (at least I purchased it after the Lexar and paid more for it). I'm going to use the Lexar before the plain Ultra, as it clearly is faster.
I was surprised at the speed difference between the write and read speeds of the Lexar. Clearly it performs faster in the read mode.
I compared my results to Rob Galbraith's (http://www.robgalbraith.com/bins/multi_page.asp?cid=6007-6425) CF database and to my surprise they are in agreement with his. The Lexar 4x it quite a bit faster in his Card-to-PC test than the Sandisk Ultra. It seems that the Card-to-Camera speed is limited by the camera. The fasted card he's tested in the Card-to-PC is not a lot faster than the rest in the Card-to-Camera.
Oh yes, and I did one more subjective test while taking the images: I placed the camera into continuous mode and made sure the flash was off and took four successive pictures by holding the shutter release. I measured the time from the press of the shutter to when the CF light went out. For the plain Ultra and the Lexar cards it was an equal 32 seconds. The UltraII was 28 seconds. Since I was shooting the same scene and the file sizes were for the most part the same, I conclude that the cards perform similarly in the camera. This result also agrees wtih Galbraith's, and his tests also include the latest Extreme-Ultra-II-Ludicrous-Speed cards. The fastest card he's tested in Card-to-PC is about 9000 Kb/sec, but it's not *that* much faster than the middle of the pack in the camera. Clearly the 300D is the limiting factor, which Galbraith also concludes.
So from my little test, the Delkin can give some significant speed improvement with the right card. Also, I feel I can use Rob Galbraith's database as a pretty good indicator of card performance.