fredmiranda.com
Login

Moderated by: Fred Miranda
  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Fuji Forum | Join Upload & Sell

1              end
  

Archive 2023 · v.60 versus v.90 for X-T5? (updated 1/18/22)

  
 
jeffbuzz
Offline
• • • •
Upload & Sell: On
p.2 #1 · v.60 versus v.90 for X-T5? (updated 1/18/22)


Knowing the true sustained write speed is key. Prograde states their v60 cards have max write of 130MB/s but only meet the basic v60 spec for 60MB/s sustained write. Angelbird v60 cards have max write of 160MB/s and minimum sustained write of 140MB/s. Fuji hasn't listed any Angelbird cards on their list. But if the Prograde v60 is good enough with only 60MB/s throughput then that may be near the limit of the camera.


Jan 11, 2023 at 08:13 PM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.2 #2 · v.60 versus v.90 for X-T5? (updated 1/18/22)


jeffbuzz wrote:
Knowing the true sustained write speed is key. Prograde states their v60 cards have max write of 130MB/s but only meet the basic v60 spec for 60MB/s sustained write. Angelbird v60 cards have max write of 160MB/s and minimum sustained write of 140MB/s. Fuji hasn't listed any Angelbird cards on their list. But if the Prograde v60 is good enough with only 60MB/s throughput then that may be near the limit of the camera.


That's sort of what I'm thinking/wondering.



Jan 11, 2023 at 08:34 PM
AZ Photo
Offline
• •
Upload & Sell: Off
p.2 #3 · v.60 versus v.90 for X-T5? (updated 1/18/22)


jeffbuzz wrote:
Knowing the true sustained write speed is key. Prograde states their v60 cards have max write of 130MB/s but only meet the basic v60 spec for 60MB/s sustained write. Angelbird v60 cards have max write of 160MB/s and minimum sustained write of 140MB/s. Fuji hasn't listed any Angelbird cards on their list. But if the Prograde v60 is good enough with only 60MB/s throughput then that may be near the limit of the camera.


All Fujifilm says is that the V60 cards are compatible - it doesn't say what compatible means and certainly doesn't say using faster cards won't give better performance. I don't think we can read anything about camera speed from the sustained write speed of a compatible card.



Jan 12, 2023 at 03:11 AM
jeffbuzz
Offline
• • • •
Upload & Sell: On
p.2 #4 · v.60 versus v.90 for X-T5? (updated 1/18/22)


AZ Photo wrote:
All Fujifilm says is that the V60 cards are compatible - it doesn't say what compatible means and certainly doesn't say using faster cards won't give better performance. I don't think we can read anything about camera speed from the sustained write speed of a compatible card.


We can read enough to know that a manufacturer is not going to recommend a card they believe would limit the camera performance in any way. Lacking any real numbers from Fujifilm that's the best we've got for now.



Jan 12, 2023 at 04:27 PM
AZ Photo
Offline
• •
Upload & Sell: Off
p.2 #5 · v.60 versus v.90 for X-T5? (updated 1/18/22)


jeffbuzz wrote:
We can read enough to know that a manufacturer is not going to recommend a card they believe would limit the camera performance in any way. Lacking any real numbers from Fujifilm that's the best we've got for now.


The only comparison test we have for the X-T5 is the link shared above - but there have been other tests for the X-H2S that show the compatible cards are not all equally fast - there is no reason why a vendor wouldn’t list compatible cards at different price points - they all work but some are faster at clearing the buffer - many will be happy to save a considerable amount of money at the expense of a little more time waiting for the buffer to clear

Here is a test example showing different speeds in camera for the compatible Lexar cards:

https://alikgriffin.com/best-memory-cards-fujifilm-x-h2s/



Jan 12, 2023 at 05:12 PM
mdude85
Offline
• • • • •
Upload & Sell: Off
p.2 #6 · v.60 versus v.90 for X-T5? (updated 1/18/22)


jeffbuzz wrote:
Knowing the true sustained write speed is key. Prograde states their v60 cards have max write of 130MB/s but only meet the basic v60 spec for 60MB/s sustained write. Angelbird v60 cards have max write of 160MB/s and minimum sustained write of 140MB/s. Fuji hasn't listed any Angelbird cards on their list. But if the Prograde v60 is good enough with only 60MB/s throughput then that may be near the limit of the camera.



140 mb/sec is not the minimum sustained write speed, it's simply the sustained write speed.

Every v60 card has a minimum speed of 60 mb/sec, that's what puts it in the v60 speed class. If a card had a minimum speed of over 90 mb/sec then it would be in the v90 class.

Prograde and Angelbird cards are rated for similar sustained write speeds (130 mb/sec vs 140 mb/sec).





Jan 12, 2023 at 05:32 PM
jeffbuzz
Offline
• • • •
Upload & Sell: On
p.2 #7 · v.60 versus v.90 for X-T5? (updated 1/18/22)


mdude85 wrote:
140 mb/sec is not the minimum sustained write speed, it's simply the sustained write speed.

Every v60 card has a minimum speed of 60 mb/sec, that's what puts it in the v60 speed class. If a card had a minimum speed of over 90 mb/sec then it would be in the v90 class.

Prograde and Angelbird cards are rated for similar sustained write speeds (130 mb/sec vs 140 mb/sec).



Does Prograde state that spec anywhere on their site? I could not find it. Angelbird does state their sustained write spec.



Jan 12, 2023 at 07:13 PM
mdude85
Offline
• • • • •
Upload & Sell: Off
p.2 #8 · v.60 versus v.90 for X-T5? (updated 1/18/22)


jeffbuzz wrote:
Does Prograde state that spec anywhere on their site? I could not find it. Angelbird does state their sustained write spec.


On prograde’s website? Not sure … you can google keywords for specific websites if you want.

Basically every v60 card is rated for 8k capture. You only need a card that’s fast enough to write whatever is being captured by the sensor. It doesn’t matter if a magical card can write at a speed of 1 gb/sec, it’s not going to make the camera work faster.

All these cards, assuming they’re about the same price, same format, and same video class are kind of interchangeable. The rest is mostly marketing.



Jan 12, 2023 at 09:33 PM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.2 #9 · v.60 versus v.90 for X-T5? (updated 1/18/22)


I finally decided to go with a v60 card (two of them actually) from the Fujifilm page listing cards that they tested with the XT5. I got a pair of 256GB Lexar "Professional"/Gold 1800x UDS-II V60 U3 Class 10 cards listed as Read speed 270MB/s and write speed 180MB/s. (They were on sale at B&H for just under $200/pair.)

Since I had my old cards, too, I decide to do a rough, unscientific test to see what differences would show up. The old cards are 128GB Lexar "Professional"/Gold 1000x UDS-II Class 10 cards listed as 150MP/s, which I assume again refers to read speed. I think the write speed is listed as 75MB/s, though it may be slower according to another listing. As far as I can tell, it has no "v" rating. (This card is NOT included on the Fujifilm page listing compatible cards.)

I pointed the camera at an iPhone running a stop watch showing .01 divisions of the second. I set the camera to the 15fps mechanical shutter mode. I captured to lossless compressed raw format (since that what I normally use) and to both cards in this format (again, that's my normal practice). With the camera set to the CH (high speed burst setting) I held the shutter down though the initial burst and then several seconds beyond.

I opened the files on my computer and used the clock timies from the iPhone screen to determine how many frames were captured in the first second, how many at full speed, and what happened to the speed of capture after it slowed.

To cut to the chase, the sub-spec card and the in-spec card were essentially the same speed during the initial burst. — e.g. the specified 15fps. (The faster card may have been going a fraction of a frame per second faster, though that could be due to the inexactness of my informal test.) Both captured the same number of frames before slowing, presumably when the camera's internal buffer filled. Both then slowed to about a 5fps rate for 4 o 5 frames. Then the camera slowed in both cases. Using the non-spec'ed 1000x card the slowdown was significant — the camera captured roughly pairs of frames (the two separated by about .25 seconds) with the pairs separated by a second or more. With the faster card the rate also slowed a bit, with frames separated by about .20 to .25 seconds, and the pairs separated by slightly wider gaps of about .3 seconds. This continued, unlike the out-of-spec card where there could be seconds between pairs.

So, the main difference in this case os what happens after the buffer fills, even when comparing an in-spec card to an out-of-spec card.

Anyone want to try the same thing with a faster v90 card and let us know what happens?



Jan 18, 2023 at 03:47 PM
mdude85
Offline
• • • • •
Upload & Sell: Off
p.2 #10 · v.60 versus v.90 for X-T5? (updated 1/18/22)


Among many caveats, it's not a great idea comparing an old card to a brand new card. The average write speed of the card falls as it is used more and more.




Jan 18, 2023 at 04:37 PM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.2 #11 · v.60 versus v.90 for X-T5? (updated 1/18/22)


mdude85 wrote:
Among many caveats, it's not a great idea comparing an old card to a brand new card. The average write speed of the card falls as it is used more and more.



First, you are right that his hardly qualifies as a thorough or complete test. However...

... if we assume that the old card's performance has degraded due to prior use, then this lightweight test would confirm even more strongly that card speed (at least through V60) has little or no effect on maximum burst speed or how many images can be captured before capture slows. (At least when using the file format and burst mode settings that I used.)

So, if anyone is hoping that a "faster" card will improve burst mode speed or depth, that's not likely. (Which is what I expected.)

I'm still interesting in seeing what happens in the post-buffer-full period when a v90 card is substituted.

For my own purposes, the v60 card is going to work just fine... at half the price of the v90s.



Jan 18, 2023 at 08:50 PM
rbf_
Offline
• • •
Upload & Sell: Off
p.2 #12 · v.60 versus v.90 for X-T5? (updated 1/18/22)


gdanmitchell wrote:
First, you are right that his hardly qualifies as a thorough or complete test. However...

... if we assume that the old card's performance has degraded due to prior use, then this lightweight test would confirm even more strongly that card speed (at least through V60) has little or no effect on maximum burst speed or how many images can be captured before capture slows. (At least when using the file format and burst mode settings that I used.)

So, if anyone is hoping that a "faster" card will improve burst mode speed or depth, that's not likely. (Which is what I
...Show more

This has been an interesting thread. I'm still on v30 cards and it's been working out fine for me but maybe I should get some faster ones for the XT5. For my GFX I prefer density to speed and just had to replace a 256GB card and went for a Sandisk Extreme Pro 512GB v30 that I got for $88(it says it's listed for 200MB/s fwiw). I was debating spending more but it would have been for 1TB v30 card going for $199. I feel the expanded density choices of the v30 cards are pretty appealing and they are not available on the faster v60 and v90 cards. I don't do a lot of shooting that fills up the buffer though. I may try some faster cards on the XT5 however as I occasionally do some coastal BIF'ing in the summer.





Jan 18, 2023 at 09:56 PM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.2 #13 · v.60 versus v.90 for X-T5? (updated 1/18/22)


rbf_ wrote:
This has been an interesting thread. I'm still on v30 cards and it's been working out fine for me but maybe I should get some faster ones for the XT5. For my GFX I prefer density to speed and just had to replace a 256GB card and went for a Sandisk Extreme Pro 512GB v30 that I got for $88(it says it's listed for 200MB/s fwiw). I was debating spending more but it would have been for 1TB v30 card going for $199. I feel the expanded density choices of the v30 cards are pretty appealing and they are not
...Show more

My general sense — and I'm sure there are at least some exceptions — is that using super fast cards really isn't going to buy you much unless you are pushing the camera's burst mode or doing video formats that require more data to be shuttled to the card at high speeds.

My out-of-spec 1000x cards (which I'd guess are pretty cheap these days) work just as well as the expensive, higher-rated cards for non-burst still photography. They work almost as well for burst mode — they seem to capture as many shots (since the camera's buffer seems to be the limit) at full rated speed, only showing a deficit after the buffer fills.

I can't say for sure with the newest cameras and newest cards, but I know that in the past there was essentially no performance benefit at all from buying cards that exceeded the specifications of the camera manufacturer. Basically the cards have to be fast enough to keep up with the camera, but the camera isn't going to go any faster than its own processors/bus allow if you put in a faster card.



Jan 18, 2023 at 10:36 PM
inkinutz
Offline
• •
Upload & Sell: On
p.2 #14 · v.60 versus v.90 for X-T5? (updated 1/18/22)


So I've been pondering this exact question and came across this post....  V60 vs V90 SD card performance in my X-T5.  I just bought a Kingston V90 card, so I decided to do some testing.

I used the following 3 cards.
- 128GB Sandisk Extreme Pro 95 MB/s (my oldest of the three without any V rating at all)
- 256GB Lexar Pro  1000X 150 MB/s
- 128GB Kingston Canvas React Plus V90 300MB/s

My testing methodology isn't exact by any means, but more to see general trends in how these cards performed in my X-T5. 

- Video - 
This was the easiest to test since the X-T5 maxes out at 360 Mbps.  This equates to 45 MB/s.  Keep in mind, that's regardless of the resolution ... 4K, 4KHQ, 6.2K, the maximum bitrate the camera offers is 360 Mbps, which means the camera will not write data to the card any faster than 45MB/s.  All 3 of my cards were able to record without issue (6.2K / 29.97P / H.265 All-I 422 / 360 Mbps).  This means for video, you will not see any additional benefit in camera for cards over V60.  If you shoot lower bitrates, then slower cards will work fine for recording as well. i.e. If you're shooting 4K at 50Mbps, I don't see why old C10 SD cards that max out at 10 MB/s writes wouldn't work.

- Photo -
Testing this was more involved.  Since what does one really mean by "camera performance" anyways?  I'm sure there might be other aspects of performance a faster SD card might contribute to, but for the purposes of this test, I decided to focus on filling and emptying the buffer.  

Here's my camera setup for the tests:
Image size L 3:2 / JPEG
Film Simulation STD
No effects or adjusted settings for  Tone, Color, etc etc.
AF-S
Electronic Shutter / 13fps (didn't want to waste my clicks with MS)

Testing procedure:
Using my phone's stopwatch.
Press the shutter (shooting in high burst) and "start" on the stopwatch simultaneously
Keep the shutter depressed until I hear the shutter noise stop or stutter.
The moment I let go of the shutter, I also press "lap" on the stopwatch.
When the blinking light on the camera stops, I press "lap" again.
Log the two times from my stopwatch; Time it took to fill the buffer & Time it took to empty the buffer.
Log the number of shots taken.
Log the total file size for all the files generated.

I did the above procedure.... 
- twice writing Lossless Compressed RAW only.
- twice writing Lossless Compressed RAW + FINE JPG
- once writing Uncompressed RAW only
- once writing Fine JPG only
... and I did this for each of the 3 cards.

See results in my table below.  

My main takeaways are .... 
1) my old trusty Sandisk Extreme Pro 95 MB/s performed really really well.  
2) the V90 card performed spectacularly.  However, I only see this benefiting only those really high demand shooters that need to burst frequently.  When shooting RAW, you'll only get a few extra frames before the buffer fills up.  But, and this is a biggy, that buffer will empty out nearly 3 times faster than the V60 card.  This will allow you to do a full burst again much sooner.    
3) Look at the V90 card JPG shots.  I had to do it twice just to be sure.  Over 400 shots before the shutter started to stutter!

I hope this will help someone with their V60 vs V90 SD card decision for their X-T5.  I'm sure I missed something somewhere.  So if there's something wrong, or you guys have any questions, do let me know!

https://www.dannyink.com/FM_Photos/buffer.png


https://www.dannyink.com/FM_Photos/3SD.jpg



Jul 22, 2023 at 03:08 AM
gyoung143
Offline
• • •
Upload & Sell: Off
p.2 #15 · v.60 versus v.90 for X-T5? (updated 1/18/22)



inkinutz wrote:

So I've been pondering this exact question and came across this post....  V60 vs V90 SD card performance in my X-T5.  I just bought a Kingston V90 card, so I decided to do some testing.

I used the following 3 cards.
- 128GB Sandisk Extreme Pro 95 MB/s (my oldest of the three without any V rating at all)
- 256GB Lexar Pro  1000X 150 MB/s
- 128GB Kingston Canvas React Plus V90 300MB/s

My testing methodology isn't exact by any means, but more to see general trends in how these cards performed in my X-T5. 

- Video - 
This was the easiest to test since the X-T5
...Show more

Thanks for that, useful info.
It seems to me that for still picture operation the important parameter is write speed, and it is not easy to find that out. Not sure what Video ratings will have to do with that.
I did some similar tests 10 years ago with my Nikon D7000, which was (and is) dismal when it came to bursts for sports photography (I remember roughly a dozen compressed RAW to fill the buffer and best part of a minute to clear the buffer) with the cards then available.
Since buying the Xpro2 about 6 years ago I've been using those good old SanDisk 95mbs cards that you have. I also tried the 170mbs versions, which seem to have no better write speed, and the Xpro2 didn't like them, but they work OK in my Xt3. These cards work ok for anything I normally do. These are only rated V30 but happily deal with my family 4k video.
When I bought the Xt3 I got a pair of the UHS-II '300mbs' SanDisk cards, they do clear the buffer much quicker, but I haven't needed to use them much as the Xt3 has a big enough buffer to hold all that I normally do. As far as I can find out these cards have a write speed of 'up to 250mbs'. Presumably if the write speed will equal the max speed the camera can generate (somebody above mentioned 360mbs?) then the buffer could be kept below saturation.

Gerry



Jul 22, 2023 at 09:32 AM
AZ Photo
Offline
• •
Upload & Sell: Off
p.2 #16 · v.60 versus v.90 for X-T5? (updated 1/18/22)



These are only rated V30 but happily deal with my family 4k video.
When I bought the Xt3 I got a pair of the UHS-II '300mbs' SanDisk cards, they do clear the buffer much quicker, but I haven't needed to use them much as the Xt3 has a big enough buffer to hold all that I normally do. As far as I can find out these cards have a write speed of 'up to 250mbs'. Presumably if the write speed will equal the max speed the camera can generate (somebody above mentioned 360mbs?) then the buffer could be kept below
...Show more

I think the 360 Mbps was for video (which is bits not bytes - so would be 45 megabytes/second). The different units, confusing rating codes, using read rather than write speeds and the difference between decimal and binary megabytes makes it all very difficult.



Jul 22, 2023 at 12:16 PM
1              end




FM Forums | Fuji Forum | Join Upload & Sell

1              end
    
 

Welcome back
Log in to your account