Timings & File Sizes

Some elements of the EOS 30D's performance are improved compared to the EOS 20D. These include startup and off to shot times and continuous shooting 'buffer depth'. The now instant power-on time appears to be achieved by the camera always being in a 'deep sleep' state and hence no requirement to go through a startup sequence or check the storage card (this happens when you change the card, camera on or not). Otherwise performance is identical to the EOS 20D. Of course that's no bad thing, the EOS 20D wasn't exactly considered slow, the EOS 30D feels snappy and 'always ready' thanks to careful use of background processing and buffering.

Timing Notes: All times calculated as an average of three operations. Unless otherwise stated all timings were made on a 3504 x 2336 JPEG Fine (approx. 2,900 KB per image).

The media used for these tests were:

  • 1 GB SanDisk Extreme III CF card
  • 1 GB Lexar Pro 133x CF card
  • 4 GB SanDisk Ultra II CF card
Time, secs
(1 GB SanDisk)
Time, secs
(1 GB Lexar)
Time, secs
(8 GB SanDisk)
Power Off to On *1   0.0
Power Off to Shot *1   <0.2
Sleep to On   0.0
Power On to Off   0.0
Record Review *2
1.3 1.3 1.4
Record Review *2
1.0 1.0 1.0
Play *3
0.8 / <0.2 0.8 / <0.2 0.8 / <0.2
Play *3
1.1 / <0.2 1.1 / <0.2 1.3 / <0.2
Play Image to Image
0.5 0.5 0.5
Play Image to Image
1.0 1.0 1.0

*1 The EOS 30D appears to achieve this by never actually being 'off', perhaps a 'deep sleep' state. If you insert a new CF card when the camera is off the CF access lamp flickers indicating that the camera is reading / checking available space.
*2 Time taken from the shutter release being pressed to the review image being displayed on the LCD monitor.

Just like other Canon digital SLRs the EOS 30D caches images which have been viewed recently to speed up browsing in play mode. The first timing is for the camera to load the image from the media card (if it has not already been cached), the second is if they have been viewed and cached by the camera.

Continuous Drive mode

To test continuous mode the camera had the following settings: Manual Focus, Manual Exposure (1/200 sec, F5.6), ISO 200. Measurements were taken from audio recordings of the tests. Media used were the same as above.

Next burst (9 indicated)

The EOS 30D uses "smart buffering" this means that it takes RAW data from the buffer, converts it to the output format (JPG or compressed CR2) and places that back in the buffer ready to be written. All of this happens in parallel, which means that space can be made in the buffer for another burst before all the images have been fully written. This exact point in time is indicated by a '9' on the buffer space display on the viewfinder LCD status bar.

The tests carried out below measured the following results for JPEG and RAW:

  • Frame rate - Initial frame rate, this was always 4.87 fps (+/- 0.02 fps)
  • Number of frames - Number of frames in a burst
  • Buffer full rate - Frame rate if shutter release held down after burst (buffer full)
  • Next burst - How soon after the camera indicates buffer space '9'
  • Write complete - How long after the last shot before the CF compartment light goes out

Burst of JPEG Large/Fine images

1 GB SanDisk
Extreme III CF
1 GB Lexar
Pro 133x CF
4 GB SanDisk
Ultra II CF
Frame rate 4.9 fps 4.9 fps 4.9 fps
Number of frames 39 40 35
Buffer full rate 1.7 fps 1.9 fps (irregular) 1 every 1.2 sec
Next burst (9 indicated) 5.2 sec 4.3 sec 9.9 sec
Write complete 12.9 sec 10.7 sec 25.4 sec

Burst of RAW images

1 GB SanDisk
Extreme III CF
2 GB Lexar
Pro 80x CF
4 GB SanDisk
Ultra II CF
Frame rate 4.9 fps 4.9 fps 4.9 fps
Number of frames 11 11 10
Buffer full rate 1 every 1.1 sec 1 every 1.4 sec 1 every 1.6 sec
Next burst (9 indicated) 8.8 sec 10.0 sec 15.3 sec
Write complete 10.5 sec 10.0 sec 16.9 sec

First of all we were curious to see the EOS 30D's actual continuous shooting rate is closer to 4.9 fps than 5.0 fps (not that the 0.1 fps will make that much difference). Next there's a noticeable improvement in buffering over the EOS 20D, the 30D manages 39 JPEG frames in a burst (25 on the 20D) and 11 RAW frames in a burst (6 on the 20D). The Nikon D200 achieves the same 'buffer depth' for JPEG but manages twice as many RAW files in a single burst.

File Flush Timing

Timings shown below are the time taken for the camera to process and "flush" the image out to the storage card. Timing was taken from the instant the shutter release was pressed to the time the storage card activity indicator beside the storage compartment went out. The activity indicator light comes almost as soon as you press the shutter release, this either means that the EOS 30D begins writing immediately or that Canon is masking the delay to write. Writing continues 'in the background' and doesn't affect any camera function. Media used were the same as above.

Image type
Time, secs
(1 GB SanDisk)
Time, secs
(1 GB Lexar)
Time, secs
(4 GB SanDisk)
3504 x 2336 RAW + JPEG *1 2.4 2.4 3.3 10,600 KB
3504 x 2336 RAW 1.9 1.8 2.4 7,700 KB
3504 x 2336 JPEG Fine 1.1 1.1 1.8 2,900 KB
3504 x 2336 JPEG Standard 1.0 1.0 1.5 1,400 KB

*1 File size reported here is the size of the RAW and JPEG files added together.
*2 The EOS 30D begins writing approximately 0.3 seconds after the shutter release is pressed so you must subtract 'processing time' from the timing to get the actual write time.

For individual shots these write times are short enough to be almost unnoticeable. Taking around two seconds to write a RAW or a second to write a Large/Fine JPEG means no waiting around. Using a slightly slower card which requires FAT32 (the 4 GB SanDisk Ultra II) had some performance impact but again not enough to make a big difference in single shoot situations.

Referring to the graphs below; with the same card the EOS 30D's throughput appears to be the same as the EOS 20D. The use of the faster SanDisk Extreme III and Lexar Pro 133x cards here show some improvement over the Ultra II cards we used in our EOS 20D review although it's not a huge advantage. These throughput figures are good but not class leading, that honor goes to the Olympus E-330 (10 MB/sec for RAW) and the Nikon D200 (8 MB/sec for RAW).

Card performance: JPEG Fine continuous burst write

Card Canon EOS 30D write speed (JPEG Fine files)
1 GB SanDisk Extreme III 5,496 KB/sec
1 GB Lexar Pro 133x 6,155 KB/sec
4 GB SanDisk Ultra II 3,435 KB/sec

Card performance: RAW continuous burst write

Card Canon EOS 30D write speed (RAW files)
1 GB SanDisk Extreme III 6,834 KB/sec
1 GB Lexar Pro 133x 7,043 KB/sec
4 GB SanDisk Ultra II 4,716 KB/sec

Cardbus 32-bit Adapter benchmark

Card Cardbus 32-bit Adapter, write speed (12.8 MP RAW files)
1 GB SanDisk Extreme III 9,368 KB/sec
1 GB Lexar Pro 133x 9,043 KB/sec
4 GB SanDisk Ultra II 4,816 KB/sec

USB transfer speed

To test the EOS 30D's USB transfer speed we transferred approximately 128 MB of images (mixed RAW and JPEG) from a SanDisk Extreme III 1 GB CF card (the same card used in the other readers). The transfer had to be carried out via WIA because Canon doesn't provide a simple 'mass storage device' feature in the camera. Thankfully this time RAW files did show up through WIA and we did get fairly typical USB 2.0 transfer rate of around 3.6 MB/sec (around 35 seconds for the 128 MB). This was about the same as our USB 2.0 card reader but not as good as a decent Firewire reader or better still a CardBus 32 PCMCIA adapter (laptop / notebook only).

Transfer rate
USB 2.0 Card reader 3.4 MB/sec
EOS 30D USB 2.0 via WIA 3.6 MB/sec
Lexar Pro Firewire Card reader 6.9 MB/sec
CardBus 32 PCMCIA adapter 11.1 MB/sec

Battery life

The EOS 30D uses the compact Canon Lithium-Ion BP-511A battery pack. This provides 1390 mAh at 7.4 V (10.3 Wh). In practice we found ourselves needing to charge the battery perhaps once in full two days of shooting. Using the CIPA battery testing method Canon's specify battery life as:

Temperature No Flash 50% Flash usage
At 20°C / 38°F 1100 750
At 0°C / 32°F 900 600