GH6 card speed

[jrsforums]

Just for the heck of it, I decided to see how fast the buffer cleared with the 3 cards I had.
ProGrade CFx 1700 read/ 1400 write

Angelbird CFx 1785r/1550w(min.1300w)

Sony ‘G’ SD 300r/299w

Test was to us HS60, start timer when shutter pressed and held until buffer full (200 shots), and stop timer when card indicated it had stopped writing. Images were RAW.

Angelbird was 40 seconds

ProGrade was 60 seconds

Sony SD card was 53 seconds

Observations

While the primary read/write specs of the CF cards were close, the Angelbird stated a minimum sustained speed and the Prograde did not. Obviously, the Prograde sustained rate is much slower than the Angelbird.

The Sony SD card, while spec’d much slower than the CF cards, performed quite well in writing RAW images.

The Sd performance surprised me quite a bit. I was wondering if this is a function of writing stills vs. video? Does Panasonic need to improve CFx speed?

 

[alcelc]

Interesting result.

Thank you for the info.

 

[FoxAdriano]

Hey friends, I have so many Sony cards I attach and I'd like to know: if I will use these cards with the Panasonic GH6, what precise problems could I get if I like to shoot at 4K 50p 10bit? Thanks for your pleasant info.

--








A lonesome traveler looking for lost tribes around the world: https://www.youtube.com/channel/UCdvI0LGfPrcIJZiWDrZCaaw/videos

 

[olrett]

Hey friends, I have so many Sony cards I attach and I'd like to know: if I will use these cards with the Panasonic GH6, what precise problems could I get if I like to shoot at 4K 50p 10bit? Thanks for your pleasant info.

See here:

Loading…

www.dpreview.com

 

[FoxAdriano]

Hey friends, I have so many Sony cards I attach and I'd like to know: if I will use these cards with the Panasonic GH6, what precise problems could I get if I like to shoot at 4K 50p 10bit? Thanks for your pleasant info.

See here:

https://www.dpreview.com/articles/6520830671/a-beginners-guide-to-memory-cards/2

My english is not good and I don't understand well if I can use my cards, even if they are slower. Anyway I thank you.

 

[Jimmy G]

Just for the heck of it, I decided to see how fast the buffer cleared with the 3 cards I had.
ProGrade CFx 1700 read/ 1400 write

Angelbird CFx 1785r/1550w(min.1300w)

Sony ‘G’ SD 300r/299w

Test was to us HS60, start timer when shutter pressed and held until buffer full (200 shots), and stop timer when card indicated it had stopped writing. Images were RAW.

Angelbird was 40 seconds

ProGrade was 60 seconds

Sony SD card was 53 seconds

Observations

While the primary read/write specs of the CF cards were close, the Angelbird stated a minimum sustained speed and the Prograde did not. Obviously, the Prograde sustained rate is much slower than the Angelbird.

The Sony SD card, while spec’d much slower than the CF cards, performed quite well in writing RAW images.

The Sd performance surprised me quite a bit. I was wondering if this is a function of writing stills vs. video? Does Panasonic need to improve CFx speed?

Hrmm, your post does not quite provide complete information...

Regarding the AngelBird CFExpress cards, folks shopping for cards will certainly wish to take a look at their individual product pages as their advertised Sustained Write speeds are all over the map depending on the size and class of the card...

CFexpress™ 2.0 | Type B | Angelbird

https://www.angelbird.com/category/cfexpresstm-20-type-b-238/

Regarding the ProGrade CFExpress cards, your information that they do not provide Sustained Write speeds is just plain erroneous and, again folks shopping for cards would do well to check the specs for their Gold and Cobalt cards where they will find the following easy-to-read chart on each product page...

ProGrade-Comp-Chart_A.jpg (WEBP Image, 970 × 600 pixels)

...a little homework goes a long way.

From your above posted test results it would seem that you tested using their Gold card (of unspecified capacity) where their Sustained Write speeds are ≤400MBs depending on capacity? Your conclusions seem to disparage ProGrade unnecessarily and methinks a retest with with a Cobalt card with a Sustained Write speed of 1400MBs would have clearly yielded a different conclusion.

Also, FWIW, I did speak with their tech support folks last week (prior to my purchasing their Cobalt 325GB card) to confirm their testing with the Panasonic GH6 (as it is currently not shown on this chart) and they informed that it had "qualified" and that they would be updating this chart, er, "shortly".

Also, Panasonic is providing a page with their own test results for both SD and CFExpress cards for the GH6 here...

Compatibilities of Memory card | Compatibility | Digital Camera | Digital AV | Support | Panasonic Global

https://av.jpn.support.panasonic.com/support/global/cs/dsc/connect/sd/dc_gh6.html

___________________

Best practices...shopping for the correct card for one's application requires some user homework, check with the manufacturer before making any decisions.

 

[pannumon]

Hey friends, I have so many Sony cards I attach and I'd like to know: if I will use these cards with the Panasonic GH6, what precise problems could I get if I like to shoot at 4K 50p 10bit? Thanks for your pleasant info.

See here:

https://www.dpreview.com/articles/6520830671/a-beginners-guide-to-memory-cards/2

My english is not good and I don't understand well if I can use my cards, even if they are slower. Anyway I thank you.

Hi!

First or all, you can always try!

Your card is UHS-I / U3. UHS-I is an older interface (UHS-II are usually expensive). U3 means it is guaranteed to write at least 30MB/s. 30MB/s is about 240mbit/s. This means you can record video at 240Mbit/s or lower bitrate. Higher bitrates may be possible.

240Mbit/s is enough for 10-bit 4k 50p, 4:2:0, LongGOP, H.265: https://www.dpreview.com/reviews/panasonic-lumix-dc-gh6-initial-review

All options except MP4 seem to be 10-bit.

On G9 UHS-I limits the write speed to around 30MB/s because the two slots share the bandwidth. I think the UHS-II limitation on G9 is about 100MB/s. GH6 can probably do better, but there may be severe limitations if works anything like G9. It's good to see the test results.

 

[pannumon]

Just for the heck of it, I decided to see how fast the buffer cleared with the 3 cards I had.
ProGrade CFx 1700 read/ 1400 write

Angelbird CFx 1785r/1550w(min.1300w)

Sony ‘G’ SD 300r/299w

Test was to us HS60, start timer when shutter pressed and held until buffer full (200 shots), and stop timer when card indicated it had stopped writing. Images were RAW.

Angelbird was 40 seconds

ProGrade was 60 seconds

Sony SD card was 53 seconds

Thanks for the data!

How much does one raw shot on average take space? From some samples it seems like 43MB, right?

With a faster card, shouldn't the camera take more shots before the buffer gets full (because the buffer is being cleared simultaneously)? This is not of course a big deal if the buffer fills up almost instantly.

If we assume 43MB/shot and 200 shots, we have 8600MB of data. Let's assume 30% uncertainty so that people understand to take these with a grain of salt. For write speeds we get:

Angelbird: 8600MB / 40s = 215MB/s +/-30%

ProGrade: 8600MB / 60s = 143 MB/s +/-30%

Sony: 8600MB / 53s = 162MB/s +/-30%

 

[jrsforums]

Hey friends, I have so many Sony cards I attach and I'd like to know: if I will use these cards with the Panasonic GH6, what precise problems could I get if I like to shoot at 4K 50p 10bit? Thanks for your pleasant info.

First, you will probably have to avoid All-Intra (see below). Then, it will depend on the speed of the Sony cards. ‘G’ cards have 299 write speed; ‘M’ cards 150 write speed.

 

[jrsforums]

Just for the heck of it, I decided to see how fast the buffer cleared with the 3 cards I had.
ProGrade CFx 1700 read/ 1400 write

Angelbird CFx 1785r/1550w(min.1300w)

Sony ‘G’ SD 300r/299w

Test was to us HS60, start timer when shutter pressed and held until buffer full (200 shots), and stop timer when card indicated it had stopped writing. Images were RAW.

Angelbird was 40 seconds

ProGrade was 60 seconds

Sony SD card was 53 seconds

Thanks for the data!

How much does one raw shot on average take space? From some samples it seems like 43MB, right?

With a faster card, shouldn't the camera take more shots before the buffer gets full (because the buffer is being cleared simultaneously)? This is not of course a big deal if the buffer fills up almost instantly.

I can only comment by what I see on the camera screen. Based on that, it appears that in ‘SH’ speed modes, the buffer fills before the cards are written to. That is, the card does not turn color (indicating it is being written to) until the counter reaches ‘0’ and shooting stops.

If we assume 43MB/shot and 200 shots, we have 8600MB of data. Let's assume 30% uncertainty so that people understand to take these with a grain of salt. For write speeds we get:

Angelbird: 8600MB / 40s = 215MB/s +/-30%

ProGrade: 8600MB / 60s = 143 MB/s +/-30%

Sony: 8600MB / 53s = 162MB/s +/-30%

As I said, some of the time was shot time, not just write time. I did not try to time from the end of shooting to buffer clear. I may get time to do that, but suspect it would show a slightly greater discrepancy in speed rates (as the constant shot rate would not be included).

 

[jrsforums]

Just for the heck of it, I decided to see how fast the buffer cleared with the 3 cards I had.
ProGrade CFx 1700 read/ 1400 write

Angelbird CFx 1785r/1550w(min.1300w)

Sony ‘G’ SD 300r/299w

Test was to us HS60, start timer when shutter pressed and held until buffer full (200 shots), and stop timer when card indicated it had stopped writing. Images were RAW.

Angelbird was 40 seconds

ProGrade was 60 seconds

Sony SD card was 53 seconds

Observations

While the primary read/write specs of the CF cards were close, the Angelbird stated a minimum sustained speed and the Prograde did not. Obviously, the Prograde sustained rate is much slower than the Angelbird.

The Sony SD card, while spec’d much slower than the CF cards, performed quite well in writing RAW images.

The Sd performance surprised me quite a bit. I was wondering if this is a function of writing stills vs. video? Does Panasonic need to improve CFx speed?

Hrmm, your post does not quite provide complete information...

Regarding the AngelBird CFExpress cards, folks shopping for cards will certainly wish to take a look at their individual product pages as their advertised Sustained Write speeds are all over the map depending on the size and class of the card...

CFexpress™ 2.0 | Type B | Angelbird

https://www.angelbird.com/category/cfexpresstm-20-type-b-238/

Regarding the ProGrade CFExpress cards, your information that they do not provide Sustained Write speeds is just plain erroneous and, again folks shopping for cards would do well to check the specs for their Gold and Cobalt cards where they will find the following easy-to-read chart on each product page...

ProGrade-Comp-Chart_A.jpg (WEBP Image, 970 × 600 pixels)

...a little homework goes a long way.

From your above posted test results it would seem that you tested using their Gold card (of unspecified capacity) where their Sustained Write speeds are ≤400MBs depending on capacity? Your conclusions seem to disparage ProGrade unnecessarily and methinks a retest with with a Cobalt card with a Sustained Write speed of 1400MBs would have clearly yielded a different conclusion.

Also, FWIW, I did speak with their tech support folks last week (prior to my purchasing their Cobalt 325GB card) to confirm their testing with the Panasonic GH6 (as it is currently not shown on this chart) and they informed that it had "qualified" and that they would be updating this chart, er, "shortly".

Also, Panasonic is providing a page with their own test results for both SD and CFExpress cards for the GH6 here...

Compatibilities of Memory card | Compatibility | Digital Camera | Digital AV | Support | Panasonic Global

https://av.jpn.support.panasonic.com/support/global/cs/dsc/connect/sd/dc_gh6.html

___________________

Best practices...shopping for the correct card for one's application requires some user homework, check with the manufacturer before making any decisions.

Thanks for the info. I went by info provided on B&H site. It was a simple test and not intended to be a buying guide. BTW, while I did not provide exact names of cards, the specs I provided pretty clearly indicate which ones they were.

Perhaps you would like to test the Cobalt card. I have no intent on purchasing one.

 

[Ranlee]

CFEXpress - Sandisk 1700R 1200W = 50 Sec Seems about right compared to yours.

Ritz SD 280R 250W = 56 Seconds

Sony SD G 300R 299W Two different cards 1st =1' 4" 2nd 1' 6" Not sure why mine are slower than yours.

Lexar SD 150Mb'S 1000x - Whatever that means = 2 minutes

Sandisk Extreme Pro 95 Mb/s = 3' 15" Won't be using these in the GH6 :-D

 

[pannumon]

As I said, some of the time was shot time, not just write time. I did not try to time from the end of shooting to buffer clear. I may get time to do that, but suspect it would show a slightly greater discrepancy in speed rates (as the constant shot rate would not be included).

I agree, the difference would be slightly larger with more exact methods*. That said, these data are still very useful. We see that the writing speeds of fast CFexpress cards on GH6 are an order of magnitude lower than what is the specified card writing speed. Conclusion: The performance between fast and very fast CFexpress cards on GH6 (current firmware) is most likely negligible.

*For a more precise calculation, format card, take burst, measure time, copy the files to a computer and write down the total size of the files.

 

[jrsforums]

CFEXpress - Sandisk 1700R 1200W = 50 Sec Seems about right compared to yours.

Ritz SD 280R 250W = 56 Seconds

Sony SD G 300R 299W Two different cards 1st =1' 4" 2nd 1' 6" Not sure why mine are slower than yours.

I’ll retest. Did you format before testing?

Lexar SD 150Mb'S 1000x - Whatever that means = 2 minutes

Sandisk Extreme Pro 95 Mb/s = 3' 15" Won't be using these in the GH6 :-D

 

[jrsforums]

As I said, some of the time was shot time, not just write time. I did not try to time from the end of shooting to buffer clear. I may get time to do that, but suspect it would show a slightly greater discrepancy in speed rates (as the constant shot rate would not be included).

I agree, the difference would be slightly larger with more exact methods*. That said, these data are still very useful. We see that the writing speeds of fast CFexpress cards on GH6 are an order of magnitude lower than what is the specified card writing speed. Conclusion: The performance between fast and very fast CFexpress cards on GH6 (current firmware) is most likely negligible.

Just a guess, but it appears that the Panasonic CFx transfer rate caps out at about 600-ish.

*For a more precise calculation, format card, take burst, measure time, copy the files to a computer and write down the total size of the files.

 

[Ranlee]

CFEXpress - Sandisk 1700R 1200W = 50 Sec Seems about right compared to yours.

Ritz SD 280R 250W = 56 Seconds

Sony SD G 300R 299W Two different cards 1st =1' 4" 2nd 1' 6" Not sure why mine are slower than yours.

I’ll retest. Did you format before testing?

Lexar SD 150Mb'S 1000x - Whatever that means = 2 minutes

Sandisk Extreme Pro 95 Mb/s = 3' 15" Won't be using these in the GH6 :-D

Yes, I did reformat all the cards prior to test. I will run it again, with the Sony G's, and see how big the files that are saved are, that would make a difference I expect.

To Pannumon: The Sony G's were type II. The Sandisk 95 Mb/s is type I as you suspected.

The Lexar is type II, as is the Ritz which is labeled Video Pro.

Just in case, the CFexpress Sandisk and the Ritz SD are both 128GB. All others are 32GB's.

 

[Ranlee]

CFEXpress - Sandisk 1700R 1200W = 50 Sec Seems about right compared to yours.

Ritz SD 280R 250W = 56 Seconds

Sony SD G 300R 299W Two different cards 1st =1' 4" 2nd 1' 6" Not sure why mine are slower than yours.

I’ll retest. Did you format before testing?

Lexar SD 150Mb'S 1000x - Whatever that means = 2 minutes

Sandisk Extreme Pro 95 Mb/s = 3' 15" Won't be using these in the GH6 :-D

Ran the test again with a Sony SD G card. 1 minute 5 seconds. 200 files, 35MB's per file.

 

[jrsforums]

CFEXpress - Sandisk 1700R 1200W = 50 Sec Seems about right compared to yours.

Ritz SD 280R 250W = 56 Seconds

Sony SD G 300R 299W Two different cards 1st =1' 4" 2nd 1' 6" Not sure why mine are slower than yours.

I’ll retest. Did you format before testing?

Lexar SD 150Mb'S 1000x - Whatever that means = 2 minutes

Sandisk Extreme Pro 95 Mb/s = 3' 15" Won't be using these in the GH6 :-D

Ran the test again with a Sony SD G card. 1 minute 5 seconds. 200 files, 35MB's per file.

Ran test again on Sony ‘Tough’ G….same ~53sec. This is a relatively new card, to me, and has not been heavily used.

I also tested on an older, very heavily used non-tough Sony G….~200sec. (Both card were 128GB)



I found this very interesting and suspect it has something to do with usage of the card and what the Sd card needs to do to write to used areas. I wonder if the ‘low level format’ such as on the CFx cards would help the SD cards.

Someone with greater knowledge may be able to explain.

I may later test some other like cards.

 

[pannumon]

Ran test again on Sony ‘Tough’ G….same ~53sec. This is a relatively new card, to me, and has not been heavily used.

I also tested on an older, very heavily used non-tough Sony G….~200sec. (Both card were 128GB)

I found this very interesting and suspect it has something to do with usage of the card and what the Sd card needs to do to write to used areas. I wonder if the ‘low level format’ such as on the CFx cards would help the SD cards.

Someone with greater knowledge may be able to explain.

I may later test some other like cards.

Try SD Memory Card Formatter from SD Association. https://www.sdcard.org/downloads/formatter/

The good old "low level" format might take a long time, but it would be extremely interesting to see if it would make a difference. To prove it, it would be great to have several test runs before and after the formatting, using the same battery, (more or less) fully charged.

 

[Ranlee]

CFEXpress - Sandisk 1700R 1200W = 50 Sec Seems about right compared to yours.

Ritz SD 280R 250W = 56 Seconds

Sony SD G 300R 299W Two different cards 1st =1' 4" 2nd 1' 6" Not sure why mine are slower than yours.

I’ll retest. Did you format before testing?

Lexar SD 150Mb'S 1000x - Whatever that means = 2 minutes

Sandisk Extreme Pro 95 Mb/s = 3' 15" Won't be using these in the GH6 :-D

Ran the test again with a Sony SD G card. 1 minute 5 seconds. 200 files, 35MB's per file.

Ran test again on Sony ‘Tough’ G….same ~53sec. This is a relatively new card, to me, and has not been heavily used.

I also tested on an older, very heavily used non-tough Sony G….~200sec. (Both card were 128GB)

I found this very interesting and suspect it has something to do with usage of the card and what the Sd card needs to do to write to used areas. I wonder if the ‘low level format’ such as on the CFx cards would help the SD cards.

Someone with greater knowledge may be able to explain.

I may later test some other like cards.

My Sony G's are probably 2 years old and have been in my G9's since purchase. The Ritz is new and in spite of lower write speed than the Sony G it tested considerably faster. Who knows?