CF Card Speeds


Status
Not open for further replies.

Hoky

New Member
Mar 17, 2004
1,182
0
0
Singapore
hoky.multiply.com
I just did a few CF read / write speed test with my cards today and I shall post the results.

Here's the test setup:

Dell Precision 380 Core Duo 300 Ghz
2 Gb RAM
WinXP Pro
Sandisk Extreme Reader (USB 2.0)
Bare minimum software operating in background (Quicktime player in taskbar and mcafee 8.0)

Test Software: SiSoftware Sandra Lite
Module: Removable software / flash hardware test.
 

Results for Ultra II 1 Gb CF

Benchmark Results
Combined Index : 2824 operation(s)/min
Endurance Factor : 28.3
Results Interpretation : Higher index values are better.

Performance Test Status
Run ID : C7ALANHO on Thursday, April 12, 2007 at 4:11:42 PM
System Timer : 3GHz

512B Files Test
Read Performance : 37620 operation(s)/min (314 kB/sec, 2x)
Write Performance : 2023 operation(s)/min (17 kB/sec, 0x)
Delete Performance : 4537 operation(s)/min
File Fragments : 1.0
Combined Index : 5369 operation(s)/min

32kB Files Test
Read Performance : 13807 operation(s)/min (7364 kB/sec, 49x)
Write Performance : 484 operation(s)/min (258 kB/sec, 1x)
Delete Performance : 769 operation(s)/min
File Fragments : 1.0
Combined Index : 1260 operation(s)/min

256kB Files Test
Read Performance : 2358 operation(s)/min (10061 kB/sec, 67x)
Write Performance : 355 operation(s)/min (1515 kB/sec, 10x)
Delete Performance : 634 operation(s)/min
File Fragments : 1.0
Combined Index : 795 operation(s)/min

2MB Files Test
Read Performance : 307 operation(s)/min (10479 kB/sec, 69x)
Write Performance : 102 operation(s)/min (3482 kB/sec, 23x)
Delete Performance : 518 operation(s)/min
File Fragments : 1.0
Combined Index : 196 operation(s)/min

64MB Files Test
Read Performance : 10 operation(s)/min (10923 kB/sec, 72x)
Write Performance : 6 operation(s)/min (6554 kB/sec, 43x)
Delete Performance : 364 operation(s)/min
File Fragments : 1.0
Combined Index : 9 operation(s)/min

Endurance Test Status
Operating System Disk Cache Used : No
Use Overlapped I/O : No
Test File Size : 32MB
Block Size : 512 byte(s)
File Fragments : 1

Endurance Benchmark Breakdown
Repeated Sector ReWrite : 162 kB/s
Sequential Sector Write : 760 kB/s
Random Sector Write : 11 kB/s

Drive
Total Size : 965MB
Free Space : 965MB, 100%
Cluster Size : 16kB

Performance Tips
Notice 5901 : CD 1x= 150kB/s; Exercise caution when comparing measured versus published ratings.
Notice 5008 : To change benchmarks, click Options.
Notice 5004 : Synthetic benchmark. May not tally with 'real-life' performance.
Notice 5006 : Only compare the results with ones obtained using the same version!
Notice 5207 : Consider using the File System Benchmark for non-Flash devices.
Notice 5900 : Endurance factor can only be used on the same type of device (SLC or MLC).
Tip 11 : Use the 'Switch Chart Type' button to switch between Detailed and Combined charts.
Tip 2 : Double-click tip or press Enter while a tip is selected for more information about the tip.
 

Results for Lexar 133x 1 Gb CF
Benchmark Results
Combined Index : 3789 operation(s)/min
Endurance Factor : 14.1
Results Interpretation : Higher index values are better.

Performance Test Status
Run ID : C7ALANHO on Thursday, April 12, 2007 at 4:30:59 PM
System Timer : 3GHz

512B Files Test
Read Performance : 39490 operation(s)/min (329 kB/sec, 2x)
Write Performance : 1856 operation(s)/min (15 kB/sec, 0x)
Delete Performance : 2857 operation(s)/min
File Fragments : 1.0
Combined Index : 4720 operation(s)/min

32kB Files Test
Read Performance : 17261 operation(s)/min (9206 kB/sec, 61x)
Write Performance : 1763 operation(s)/min (940 kB/sec, 6x)
Delete Performance : 2895 operation(s)/min
File Fragments : 1.0
Combined Index : 4176 operation(s)/min

256kB Files Test
Read Performance : 3382 operation(s)/min (14430 kB/sec, 96x)
Write Performance : 1235 operation(s)/min (5269 kB/sec, 35x)
Delete Performance : 2926 operation(s)/min
File Fragments : 1.0
Combined Index : 2200 operation(s)/min

2MB Files Test
Read Performance : 441 operation(s)/min (15053 kB/sec, 100x)
Write Performance : 259 operation(s)/min (8841 kB/sec, 58x)
Delete Performance : 2899 operation(s)/min
File Fragments : 1.0
Combined Index : 392 operation(s)/min

64MB Files Test
Read Performance : 14 operation(s)/min (15292 kB/sec, 101x)
Write Performance : 13 operation(s)/min (14199 kB/sec, 94x)
Delete Performance : 1323 operation(s)/min
File Fragments : 1.0
Combined Index : 15 operation(s)/min

Endurance Test Status
Operating System Disk Cache Used : No
Use Overlapped I/O : No
Test File Size : 32MB
Block Size : 512 byte(s)
File Fragments : 1

Endurance Benchmark Breakdown
Repeated Sector ReWrite : 112 kB/s
Sequential Sector Write : 796 kB/s
Random Sector Write : 22 kB/s

Drive
Total Size : 965MB
Free Space : 965MB, 100%
Cluster Size : 16kB

Performance Tips
Notice 5901 : CD 1x= 150kB/s; Exercise caution when comparing measured versus published ratings.
Notice 5008 : To change benchmarks, click Options.
Notice 5004 : Synthetic benchmark. May not tally with 'real-life' performance.
Notice 5006 : Only compare the results with ones obtained using the same version!
Notice 5207 : Consider using the File System Benchmark for non-Flash devices.
Notice 5900 : Endurance factor can only be used on the same type of device (SLC or MLC).
Tip 11 : Use the 'Switch Chart Type' button to switch between Detailed and Combined charts.
Tip 2 : Double-click tip or press Enter while a tip is selected for more information about the tip.
 

Results for Sandisk Extreme 4 Gb CF

Benchmark Results
Combined Index : 1871 operation(s)/min
Endurance Factor : 23.4
Results Interpretation : Higher index values are better.

Performance Test Status
Run ID : C7ALANHO on Thursday, April 12, 2007 at 4:35:43 PM
System Timer : 3GHz

512B Files Test
Read Performance : 27329 operation(s)/min (228 kB/sec, 1x)
Write Performance : 794 operation(s)/min (6775 bytes/sec)
Delete Performance : 3111 operation(s)/min
File Fragments : 1.0
Combined Index : 2315 operation(s)/min

32kB Files Test
Read Performance : 12504 operation(s)/min (6669 kB/sec, 44x)
Write Performance : 687 operation(s)/min (366 kB/sec, 2x)
Delete Performance : 3202 operation(s)/min
File Fragments : 1.0
Combined Index : 1938 operation(s)/min

256kB Files Test
Read Performance : 3571 operation(s)/min (15236 kB/sec, 101x)
Write Performance : 588 operation(s)/min (2509 kB/sec, 16x)
Delete Performance : 3250 operation(s)/min
File Fragments : 1.0
Combined Index : 1410 operation(s)/min

2MB Files Test
Read Performance : 517 operation(s)/min (17647 kB/sec, 117x)
Write Performance : 158 operation(s)/min (5393 kB/sec, 35x)
Delete Performance : 1879 operation(s)/min
File Fragments : 1.0
Combined Index : 321 operation(s)/min

64MB Files Test
Read Performance : 16 operation(s)/min (17476 kB/sec, 116x)
Write Performance : 14 operation(s)/min (15292 kB/sec, 101x)
Delete Performance : 535 operation(s)/min
File Fragments : 1.0
Combined Index : 17 operation(s)/min

Endurance Test Status
Operating System Disk Cache Used : No
Use Overlapped I/O : No
Test File Size : 32MB
Block Size : 512 byte(s)
File Fragments : 1

Endurance Benchmark Breakdown
Repeated Sector ReWrite : 802 kB/s
Sequential Sector Write : 790 kB/s
Random Sector Write : 23 kB/s

Drive
Total Size : 3.8GB
Free Space : 3.8GB, 100%
Cluster Size : 4kB

Performance Tips
Notice 5901 : CD 1x= 150kB/s; Exercise caution when comparing measured versus published ratings.
Notice 5008 : To change benchmarks, click Options.
Notice 5004 : Synthetic benchmark. May not tally with 'real-life' performance.
Notice 5006 : Only compare the results with ones obtained using the same version!
Notice 5207 : Consider using the File System Benchmark for non-Flash devices.
Notice 5900 : Endurance factor can only be used on the same type of device (SLC or MLC).
Tip 11 : Use the 'Switch Chart Type' button to switch between Detailed and Combined charts.
Tip 2 : Double-click tip or press Enter while a tip is selected for more information about the tip.
 

By the way, the last card tested was Extreme III.
Software used was SiSoftware Sandra Lite XI.SP1a.
 

Here are graphical summary comparing the three cards to HDD to a generic test data.

456337300_6dd3100f10.jpg


456337302_a69e4132f2.jpg


Legend:

1. Red: Sata HDD
2. Orange: Sandisk Ultra II 1 Gb CF
3. Green: Lexar 133x Professional 1 Gb CF
4. Blue: Sandisk Extreme III 4Gb CF
5. Purple: Saved data from S/W on Toshiba micro SD 256mb
 

nice test... maybe I should get myself a extreme reader as well. ;p
 

Wow! I love your setup! ;)

Thanks for the benchmarks!

Here's the test setup:

Dell Precision 380 Core Duo 300 Ghz
2 Gb RAM
WinXP Pro
Sandisk Extreme Reader (USB 2.0)
Bare minimum software operating in background (Quicktime player in taskbar and mcafee 8.0)
 

i dun understand....which is the best?
 

so the extreme IV is not as great as it seems?:dunno: cant read all the data for nuts...
 

so the extreme IV is not as great as it seems?:dunno: cant read all the data for nuts...

the test was done on an Extreme III card, not IV.. III is around 80x IIRC, so naturally the lexar would be faster...

Extreme IV is rated at 40MB/s, which is twice the rate of Extreme III (20MB/s).. so u should see a significant difference when using the Extreme IV, assuming ur writers/readers can read/write at that speed...

Exreme IV cannot be tested accurately on the TS's card-reader, as the reader is rated at only 20MB/s, to match the extreme III card.. If u want to get the Extreme IV u'll have to try the Sandisk/Lexar Firewire card readers...
 

the test was done on an Extreme III card, not IV.. III is around 80x IIRC, so naturally the lexar would be faster...

Extreme IV is rated at 40MB/s, which is twice the rate of Extreme III (20MB/s).. so u should see a significant difference when using the Extreme IV, assuming ur writers/readers can read/write at that speed...

Exreme IV cannot be tested accurately on the TS's card-reader, as the reader is rated at only 20MB/s, to match the extreme III card.. If u want to get the Extreme IV u'll have to try the Sandisk/Lexar Firewire card readers...
I believe extreme III is rated either 120x or 133x depending on SD or CF... thats why it is rated 20Mb/s (150Kb/s x 133x = 20Mb/s)
 

I believe extreme III is rated either 120x or 133x depending on SD or CF... thats why it is rated 20Mb/s (150Kb/s x 133x = 20Mb/s)

oops... yeah.. my bad.. thanks for pointing it out... Extreme III is technically at 133x...

but anyway the difference isn't that significant, and the difference is much less at larger files, so i don't think it's of any concern.. to me, either card would do...
 

Referring to the graph... the HDD is the best. :bsmilie:
But for CF speeds, the Lexar 133X seems to "outrun" the rest.
One question, how much performance advantage does the lexar enjoys when I am shooting?

To me, how fast I can doewnload the files from the card to the HDD is not very revelent.

How fast can the camera write to the CF Card when the buffer is full so that there is sufficient memory buffer for me to take the next shot is more important.
 

One question, how much performance advantage does the lexar enjoys when I am shooting?

To me, how fast I can doewnload the files from the card to the HDD is not very revelent.

How fast can the camera write to the CF Card when the buffer is full so that there is sufficient memory buffer for me to take the next shot is more important.

The tests do perform and show results of writing of read / write (and rewrite of data).
And I agree, the data bus of the camera to the CF is indeed typically the lag and not the card storage.

If you study the data I posted earlier, it may seems that the cards "write slower" at smaller files. This is due to slower card initialization and response. That is the crucial factor that we need to base in order benchmark the cards. Esp that our files may vary between 3 to 16mb. (taking 2mb results as a baseline)

OT a little:
I did a test on a friend's transcend 2Gb 120x CF... way freaking slow for the small files write. Worst than my Hagiwara unrated 512mb CF. Sorry I don't have data shown as it was done in a hurry prior to knocking off.
 

One last thing to add.
Lexar also have WA (Write acceleration) technology which is not being embedded in my reader / writer.

So I should expect faster write results w/ the DSLRs (e.g. Nikon) over the Sandisk?
 

Status
Not open for further replies.