Hey!
I just have bought budget Canon EOS M with a Canon 22 f/2 lens from a co-worker who's working for the marketing department of Apknite. I'm a comp/dmp adept and my main reason for buying this equipment was to take some photo references for matte painting and to record some short clips to exercise roto/paint and other comp stuff. As I learned, EOS M was one of the cheapest gear at the market that offers to record in RAW, and yes, I'm aware it's not a pro camera...not even close.... yet it's still better than smartphone:) Since I'm a total newbie here, I would like to ask you about the best option in terms of SD/SDHC card for my EOS M to install Magic Lantern and record/take photos in RAW.
Extra questions, do you have any experience with small, portable, not expensive tripods? Which one is worth buying for someone on a tight budget?
Thank you very much for any suggestions,
Cheers!
Sandisk Extreme Pro 95 MB/s (https://www.cameramemoryspeed.com/reviews/sd-cards/sandisk-extreme-pro-95mbs-uhs-i-64gb-sdxc-memory-card/) or same type with higher/lower storage capacity.
Cards with faster designation (150, 170 MB/s) are a tad slower and UHS-II cards have to work in UHS-I mode when used in EOS M.
Quote from: jer3my on August 14, 2019, 10:18:04 AM
Extra questions, do you have any experience with small, portable, not expensive tripods? Which one is worth buying for someone on a tight budget?
I recommend the great ZEEK channel
https://www.youtube.com/channel/UCCn0RQHYheA41EudwmLnYKA/videos
Quote from: walter_schulz on August 16, 2019, 07:20:17 AM
Cards with faster designation (150, 170 MB/s) are a tad slower and UHS-II cards have to work in UHS-I mode when used in EOS M.
from another thread:
Quote from: Walter Schulz on March 17, 2019, 08:13:54 PM
Performance of those 300 MB/s cards (UHS-II interface) is inferior to 95 MB/s cards (UHS-I). When used in cams with UHS-I interface ... and your M doesn't have UHS-II.
Hi Walter, found a good deal on a 128gb uhs-II sandisk extreme pro sdcard, will it work with the EOS-M or do i need to get a UHS-1?
Thanks for any info
Answered in quote.
If it is an older 280 MB/s Extreme Pro: Don't use it. In compatibililty mode write performance is just around 46 MByte/s.
Sorry Walter but when reading your answers i got confused, thats why i asked, you say performance of uhs2 card is slower than uhs1 on a uhs1 interface but you finalize your sentence with saying the EOS-M doesnt have UHS2 interface...language barrier for me i guess, i just wanted to know if this uhs2 card will handle the all the high resolution recording modes on the EOS-M or will the write speed drop below 50MB/s because of the uhs1 interface?
Thanks
just read your (edited) answer, this is the card:
https://www.nanochip.pt/pt-pt/produto/sandisk-micro-sdxc-card-extreme-pro-uhs-ii-128gb-sansdsqxcy128ggn6m
when i search for partnumber "SDSQXCY-128G-GN6MA" i only get UHS 1 results, maybe its a typo or wrong image on website.
Look at these. Most use these models.
95 or 170 MB/s read speed.
https://sandisk-store.pl/pl/p/SanDisk-Extreme-Pro-SDXC-64GB-V30-U3-4K-17090-MBs-typ-SDSDXXY-064G-GN4IN-Karta-Pamieci/23078558
https://sandisk-store.pl/pl/p/SanDisk-Extreme-Pro-SDHC-32GB-V30-U3-4K-9590-MBs-typ-SDSDXXG-032G-GN4IN-Karta-Pamieci/23077962
Thanks yokashin, i was looking for a shop in my country, i just called the shop but i seems they're as confused as i was, got to look elsewhere.
Once again thanks.
Serial number and picture are for UHS-I card. Card is listed for around 33 € in german shops. Can't see why this should be a deal.
I have no information if UHS-II cards (which will work in EOS M, as I wrote in quoted texts) will do overclocking in UHS-I compatibility mode.
hi, amazon DE is 33 euros but out of stock, free delivery not available for other countries plus delivery charges, over here for 40 euros and next day delivery is a good deal, thanks Walter i guess it must be an error on the website.
https://www.cameramemoryspeed.com/reviews/sd-cards/sandisk-extreme-160mbs-v30-a2-128gb-microsdxc-memory-card/ is a far better deal and widely available.
I might have left out a SMALL detail, i have a DJI Osmo Pocket and it records video @100MB/s thats why i was looking for the extreme pro model since it supports those speeds, if i ever stop using the EOS-M i can use the card on the Osmo Pocket.
Thanks for all the advices, will wait for those results when you test uhs2 on uhs1 device.
Sorry, but I think your Osmo records at 100 MBit/s, not 100 MByte/s. And Extreme Pro with UHS-I interface won't write at 100 MByte/s ...
:) you're right so much writting variations with lower/uppercase "b", i completely forgot about that, i just saved 40 euros, the DJI Osmo records @100mbps (12.5MBps), i have a 95MBps 64GB on the Osmo when it only needs a 12.5MBps card, from now on that one goes to the EOS-M.
Thanks Walter.
New to the forum, here to post some findings. Just got my first Eos M and I have some sd/micro sd cards laying around, before I buy a new one I thought I would test them out first.
Samsung EVO select microsd 128GB, 160Mhz gives best results @ 54.5MB/s, which is enough for 1080p at 14bit, which is the only mode I need, this card runs about $20 on amazon.
Samsung EVO select microsd 256GB, 192Mhz gives best results @ 56MB/s, which is enough for 1080p at 14bit, this card currently sells at $35 on amazon, I got it when it was on sale at $25.
sandisk extreme Pro 95MB/s 32Gb, 192Mhz gives best results @ 44MB/s, just about barely enough for 1080p at 14bit.
SanDisk Ultra 64GB MicroSDXC, this card is a no no, it barely goes above 20MB/s, so not enough for magic lantern.
Patriot LX Series 64GB, slightly better than Sandisk Ultra, goes to about 28MB/s, still not enough for magic lantern.
So shooting at 1080p 16:9 at 14bit, you are gonna need at least 41 MB/S writing speed. A 64GB card is gonna give you around 20minutes of footage, 128GB 40min, 256GB 80min and so on so forth. Based on my personal experience, a 8-10 hours shooting day(story based short film), I usually get about 60-75 mins of actual footage on camera. So for me, 256GB is the minimum.
So if you are like me looking for a cheaper alternatives to the much praised but more expensive SanDisk Extreme PRO, samsung evo select is a pretty good contender, they are not as fast as Sandisk extreme pro, but good enough for what I need.
Currently I'm waiting for a 256GB Samsung pro plus full size SD card to be delivered, I got it when the price is marked down to $30 for 256GB size, not sure it's temporary or permanent. It's supposedly designed for 'professional' use so I expect it to be more durable and writes fast than the consumer grade EVO select. I will report back when I receive it.
Ok, got the Samsung Pro Plus sd card today (not the micro-sd version)
It's slightly better thatn evo select micro sd version but not by much.
160Mhz writes at 54.8MB/s
192Mhz writes at 59.7MB/s
240Mhz writes at 19.7MB/s
so 192Mhz is the best fit. And since it's marked as 'professional', hopefully it will be more durable than evo select version.
Quote from: wmtr on July 05, 2021, 08:56:58 PM
Ok, got the Samsung Pro Plus sd card today (not the micro-sd version)
It's slightly better thatn evo select micro sd version but not by much.
160Mhz writes at 54.8MB/s
192Mhz writes at 59.7MB/s
240Mhz writes at 19.7MB/s
so 192Mhz is the best fit. And since it's marked as 'professional', hopefully it will be more durable than evo select version.
I tested Samsung pro plus micro sd(with adapter) It reached out 67MB/s Writing only in 160MHz, anything beyond that falls to 19MHz.
Waiting for a SanDisk Extreme Pro I bought directly from Western digital website(hoping it is not a counterfeit card :-\)
(https://drive.google.com/file/d/1kLvQRPUKmlSKNsndm2Bdg4CpUKZ1N36A/view?usp=drivesdk)
Quote from: jdukesa on June 13, 2022, 04:22:19 AM
I tested Samsung pro plus micro sd(with adapter) It reached out 67MB/s Writing only in 160MHz, anything beyond that falls to 19MHz.
Waiting for a SanDisk Extreme Pro I bought directly from Western digital website(hoping it is not a counterfeit card :-\)
(https://drive.google.com/file/d/1kLvQRPUKmlSKNsndm2Bdg4CpUKZ1N36A/view?usp=drivesdk)
Hey! I got my EOS M a couple days ago and a Samsung Pro Plus Micro SD (with adapter) 128GB today. I can't seem to get more than 56MB/s at 160MHz. Anything above or below just drops to insanely low speeds... Any tips? I'm thinking about just putting this card on my Sister's camera and get a Sandisk Extreme Pro for myself...
Hello,
I've just bought a new recent 64GB Sandisk Extreme Pro SDXC UHS-I 200 Mb/s 90 Mb/s for my new EOS M/15-45 combo and I don't get more than 56-57 Mb/s (red and yellow) on live recording at 5k frtp 16:9, default 190 MHz, for a few seconds for dark scenes, two seconds for brighter ones. If I choose any other frequency, speed drops.
I tried manual pc format up to 512kb cluster size exfat, to no improvement.
I was expecting 70 Mb/s as I read somewhere with regular 170 Mb/s Extreme Pro. I don't know if this new card works well with ML.
We have no way to predict how a particular card type may react to overclocking. Results may differ for different camera types, too.
And it is the first time any test was done for this recently introduced card. I read about an "UHS312" mode used by Sandisk.
And, yes: Extreme Pro in its 170 MB/s variety is considered to be gold standard ATM.
@wyup
Run card benchmarks in PLAY mode:
-Set SD overclock to 192 MHz.
-Load Bench.mo from "Modules" tab in ML settings, restart camera.
-Turn OFF RAW video and crop modes.
-Go to "Debug" tab from ML settings --> Benchmarks --> Card Benchmarks --> Quick R/W benchmark (1 min) then press PLAY button.
Share the result.
I have no way to enter ML menu in PLAY mode. I can only enter ML menu by pressing down/trashcan in LiveView mode.
I've run two benchmarks as you say, in HD 1080p, Global Draw off, Raw video off and sound recording OFF.
1st test: 1st pass: W: 59.7 MB/s R: 64.3 MB/s
2nd pass: W: 60.0 MB/s R: 64.1 MB/s
2nd test: 1st pass: W: 60.0 MB/s R: 65.2 MB/s
2nd pass: W: 60.4 MB/s R: 65.6 MB/s
Quote from: wyup on June 27, 2022, 12:31:49 AM
I have no way to enter ML menu in PLAY mode.
Start benchmark in photo and video mode. Immediately after starting press "Play" button.
Quote from: Walter Schulz on June 27, 2022, 12:39:14 AM
Start benchmark in photo and video mode. Immediately after starting press "Play" button.
I can't enter ML menu from Playback mode in any photo or video mode. Even if I inmediately press "Play" button upon upon starting.
In practical recording I can't get past 57 MB/s.
Please follow instructions!
1) Start benchmark
2) Press "Play" button
Not other way round!
Quote from: wyup on June 27, 2022, 12:31:49 AM
1st test: 1st pass: W: 59.7 MB/s R: 64.3 MB/s
2nd pass: W: 60.0 MB/s R: 64.1 MB/s
BTW, these numbers matches my Sandisk Extreme Pro 64GB UHS-I U3 170MB/s at the same settings/mode.
We still need a benchmarks outside LiveView to check how fast your SD card can reach.
Sandisk Extreme Pro 64GB SDXC UHS-I 3 V30 - 200 MB/s 90 MB/s
Canon EOS M, ML 2022 May 30
HD 1080p, RAW OFF, Sound OFF, Global Draw OFF, 192 MHz
In Play Mode:
PHOTO MODE:
1st test: 1st pass: W: 76.1 MB/s R: 85.9 MB/s
2nd pass: W: 75.7 MB/s R: 85.9 MB/s
2nd test: 1st pass: W: 76.1 MB/s R: 86.0 MB/s
2nd pass: W: 75.0 MB/s R: 86.0 MB/s
VIDEO MODE:
1st test: 1st pass: W: 75.7 MB/s R: 85.9 MB/s
2nd pass: W: 75.7 MB/s R: 85.9 MB/s
2nd test: 1st pass: W: 75.2 MB/s R: 85.9 MB/s
2nd pass: W: 76.1 MB/s R: 85.9 MB/s
Okay, your SD card performs as 170MB/s version in 192 MHz, which mean you reached the maximum possible write speed when using 192 MHz overclock (your card is fine).
Back to your question, why you only get ~57 MB/s during RAW video recording?
-Enable the new hacks (lvface + aewb) from RAW video submenu, see:
https://youtu.be/a8FM_nlA8R8?t=39
Report back.
You can also use lower bit-depths (this will reduce required write speed).
Quote from: theBilalFakhouri on June 26, 2022, 08:57:54 PM
@wyup
Run card benchmarks in PLAY mode:
-Set SD overclock to 192 MHz.
-Load Bench.mo from "Modules" tab in ML settings, restart camera.
-Turn OFF RAW video and crop modes.
-Go to "Debug" tab from ML settings --> Benchmarks --> Card Benchmarks --> Quick R/W benchmark (1 min) then press PLAY button.
Share the result.
Not sure if I should've changed from 2.5k FRTP mode to anything else before pressing "Play", but HOLY that did the trick on my end... Does it mean that I should/can keep the Samsung Pro Plus 128GB or I'd still get faster speeds with a Sandisk Extreme Pro?
1st test (165hz OC): 1st pass: W: 67.9 MB/s R: 74.7 MB/s
2nd pass: W: 67.9 MB/s R: 74.7 MB/s
2nd test(165hz OC): 1st pass: W: 68.0 MB/s R: 74.6 MB/s
2nd pass: W: 67.9 MB/s R: 74.7 MB/s
EDIT: I've noticed that I can't go above W: 56MB/s and I get a malloc error when reading even if I have the "lvface + aewb" hack enabled and the "af off" set while in "record mode".
What I mean is, having RAW enabled with the hacks enabled or having raw completely disabled is the exactly same thing in terms of speed.
Why would that happen?
Thanks!
Quote from: theBilalFakhouri on June 27, 2022, 02:42:07 AM
Okay, your SD card performs as 170MB/s version in 192 MHz, which mean you reached the maximum possible write speed when using 192 MHz overclock (your card is fine).
Back to your question, why you only get ~57 MB/s during RAW video recording?
-Enable the new hacks (lvface + aewb) from RAW video submenu, see:
https://youtu.be/a8FM_nlA8R8?t=39
Report back.
You can also use lower bit-depths (this will reduce required write speed).
With (lvface + aewb) hack, benchmark speed results are the same. With 240 MHz, results drop to ~20 MB/s. With 160 MHz, results drop to 67/74 MB/s.
In practical recording, (lvface + aewb) hack enabled, 5k frtp 2.39:1 14-bit, I can't get past 59-60-61 MB/s, I can record up to ~10 secs. With 12-bit, I can get almost continous recording.
How much practical recording speed do you get with 170 MB/s Sandisk Extreme Pro?
@wyup
I didn't ask to test other overclocking presets.
Quote from: wyup on June 27, 2022, 12:41:41 PM
In practical recording, (lvface + aewb) hack enabled, 5k frtp 2.39:1 14-bit, I can't get past 59-60-61 MB/s, I can record up to ~10 secs. With 12-bit, I can get almost continous recording.
Are using 192 MHz or 160 MHz overclock here?
Just to make sure, are these speeds (59-60-61 MB/s) are coming from card benchmarks or RAW video?
These hacks will have an effect
only while recording RAW video, it
won't have an effect while in Idle.
Check the write speed reported by RAW video while recording, like this image (https://ibb.co/CMMg20Q), take a picture for camera screen using smartphone and show me the picture.
If you can record a video showing what are you doing (your camera settings), that's would be better too.
Quote from: wyup on June 27, 2022, 12:41:41 PM
How much speed do you get with 170 MB/s Sandisk Extreme Pro?
~72 MB/s while recording at the same settings (192 MHz overclock).
Quote from: theBilalFakhouri on June 27, 2022, 01:08:14 PM
@wyup
I didn't ask to test other overclocking presets.
Are using 192 MHz or 160 MHz overclock here?
Just to make sure, are these speeds (59-60-61 MB/s) are coming from card benchmarks or RAW video?
These hacks will have an effect only while recording RAW 14-bit video, it won't have an effect while in Idle.
Check the write speed reported by RAW video while recording, like this image (https://ibb.co/CMMg20Q), take a picture for camera screen using smartphone and show me the picture.
If you can record a video showing what are you doing (your camera settings), that's would be better too.
~72 MB/s while recording at the same settings (192 MHz overclock).
These speeds (59-60-61 MB/s) are coming from RAW actual video scene shooting, at standard 192 MHz, (lvface + aewb) hack, and as I said, in practical recording, not benchmarking. Of course, it is the small corner screen reading, none otherwise.
I'm trying 5k frtp, raw ON 1440x1810 1:1, Kill global Draw ON, lvface +aewb ON, ratio 2.39:1, sound recording OFF, Global Draw OFF, rest defaults.
I'm using MF, Servo OFF, Manual Exposure, ISO 100, about 1/60, F8, exterior, white wall optimal exposure.
Is there any other max bitrate mode or best setting to test?
I'm still far from 72 MB/s.
@wyup
Perform "Low level format" in camera from Canon menu, does that help? same write speed?
Yes, I did low level format from Canon menu.
What write speed you get if you disabled "More hacks" with the same settings (https://www.magiclantern.fm/forum/index.php?topic=24401.msg239437#msg239437) above?
Also, what option are you using for "Small hacks" in RAW video submenu?
Small hacks affect write speed too.
You don't get 72Mb/s in 192Mhz and 240Mhz don't work on eos m. Around 60 seems right.
If you do benchmark stay in photo mode and also push play button just before bench marking starts.
Quote from: Danne on June 28, 2022, 04:08:04 PM
You don't get 72Mb/s in 192Mhz .. . Around 60 seems right.
What makes EOS M perform worse at the same settings compared to 100D/700D?
At same the used settings (https://www.magiclantern.fm/forum/index.php?topic=24401.msg239437#msg239437), both 100D/700D get up to ~70 MB/s with 192 MHz overclock.
~10 MB/s is a big difference.
@wyup
There is a case, if your scene requires high write speed like 80 MB/s or higher when you shoot at 14-bit lossless, in this case RAW video will report ~60MB/s and recording would stop early.
Anyway, I will prepare card benchmarks test with all hacks enabled, to be run in 1080p HD mode and 5K FRTP preset (while not recording), and then let's what benchmarks numbers would say.
Sometimes my Adata writes with 80 MByte/s in 192 MHz. Not consistent, though. And benchmark, of course.
Here is Bench.mo (https://drive.google.com/file/d/1TpbW3F_g3kpzW9KqVuyBHLb7SSoDsMDD/view?usp=sharing) with all hacks enabled. Only for EOS M.
-Hacks included:
-All "Small hacks".
-lvface + aewb.
-How to:
-Copy bench.mo to ML/modules folder, replace the old bench.mo with the provided one.
-Load the module, restart camera.
-Set your crop mode preset and other settings.
-Run "Quick R/W benchmark (1 min)".
-Share the result.
-Camera might freeze in 1080p HD, iirc new hacks doesn't work in MCM, 1080p HD preset is using MCM, so if that happens, skip benchmarking in this mode, pull the battery.
-After benchmarks finish, you need to make a camera restart to turn off the hacks.
-If you ran benchmarks and you want to run new benchmarks with different settings/crop mode, restart the camera first.
Benchmark test with Bilal's 'bench.mo', from Photo mode + play button as Danne says, I get:
1st: W: 75.8 R:85.9
2nd:W: 76.0 R:86.0
Recording without modules, raw submenu> small hacks: af OFF, more hacks: lvface + aewb, Use SRM memory ON, kill global draw ON, aspect ratio: 1:1, resolution: 1360x1976, rest: 5k anamorphic frtp, sd overclock 192MHz, sound recording OFF,overlay global draw OFF, 14-bit
Average recording speed: 57-58-59 Mb/s, MF, ME, ISO 3200, bright exposure lasts for ~6 seconds.
Run benchmarks in video mode, in LiveView with this bench.mo (https://www.magiclantern.fm/forum/index.php?topic=24401.msg239450#msg239450), not Play mode.
Quote from: theBilalFakhouri on June 28, 2022, 06:57:35 PM
Run benchmarks in video mode, in LiveView with this bench.mo (https://www.magiclantern.fm/forum/index.php?topic=24401.msg239450#msg239450), not Play mode.
1st: W: 70.3 R:71.5
2nd: W: 70.9 R:70.0
+malloc errors
You can get rid of malloc errors if you turn off RAW video before running benchmarks.
Anyway, now your result matches 100D/700D with (Sandisk 170MB/s) at the same settings. RAW video recording should get up to 70 MB/s too (at least this happens on 100D/700D).
And again, your SD card seems fine.
I don't know if this an issue related to what things Danne have changed in his custom build which prevent getting 70 MB/s write speed, it could be his implementation/changes to small/new hacks (hacks don't work in some cases while recording RAW video?), or maybe other things?.
I can't help with these since I don't have an EOS M to check.
Could you check "Free memory" on EOS M?, follow this guide (https://www.magiclantern.fm/forum/index.php?topic=26521.msg239231#msg239231).
Less memory has an impact on write speed too.
Guys, I've found the fastest way to record, I manage sustained 63-64-65 Mb/s (65.7 max) in yellow, 5k frtp 1360x1976 14-bit, almost continous, and 2.8k 10-bit continous.
The trick is to format externally with lowest possible sector size. I've found out physical sector size of SD card is 512 bytes. You find out with 'fsutil fsinfo sectorinfo volume:' in Windows. From there, manual format to minimum camera bootable, which I've found is 1024 bytes. I format with this command:
FORMAT /FS:EXFAT /A:1024 /Q G:
and then copy your ML filesystem and install.
It takes some 15 seconds to boot. Also I've found that sometimes at first it stops recording, but next time it gets to full 63 MB/s.
The first sector size I tried was 4096 bytes, it booted normally and it seemed already faster. 1024 is stable enough for me and the fastest.
EDIT: This trick *seems* to freeze ML debug memory info submenu. At 1024, it takes +12 seconds to boot. At 2048, it takes some 8 seconds to boot. At 4096, the lowest selectable from windows format ui, it boots normally and speed sets to ~61 MB/s. With stock in-camera format (128k sector), recording speed drops again to 57-58 Mb/s.
I think there is something to look upon on ML software regarding low level card writing strategy, we're talking about 5-8 MB/s faster writing with smaller sector sizes.
EDIT 2: I think the improvement is with just formatted card. As long as I record, speeds seem to return to normal.
I dont know man , it boots forever ( over 30 seconds blinking green ) and boots without magic lantern, how did you set bootable flag ? I used EOSCard 1.40 but it didnt help
Try EOScard, select boxes "EOS_DEVELOP" and "BOOTDISK" and press "Save" button.
OK works, i tried 2048 , and eoscard to set bootable flag and develop...
So , it boots about 10 seconds blinking red,
OK after a bit of testing 1024, i think this is too unstable , sure yo uget some seconds but then it drops and stops , stock settings are just more stable but lets others try it out
Boot times are killing me as well, takes like 15 seconds to boot into magic lantern and with 1024 mode is like 30 secs.
OK on 2048 its more stable but then its almost as good as stock format settings in cam + stock settigs have better boot time.
Maybe this would be better to test with 2.9k mode or 3k mode and 24fps but these are not in camera.
what is the command for Mac?
sudo news_exfat -s 2 disk2s1
is that right?
BTW, I tried these (https://www.magiclantern.fm/forum/index.php?topic=24401.msg239455#msg239455) on my Sandisk Extreme PRO 64GB UHS-I U3 170 MB/s and it performed worse when I ran benchmarks (using 240 MHz overclock on 700D):
Write speed:
1024 bytes: 82 MB/s
2048 bytes: 88 MB/s
4096 bytes: 89 MB/s
8192 bytes: 90 MB/s (matches 128 KB)
Read speed:
All was ~106 MB/s, no differences here.
Quote from: theBilalFakhouri on July 02, 2022, 02:32:39 AM
BTW, I tried these (https://www.magiclantern.fm/forum/index.php?topic=24401.msg239455#msg239455) on my Sandisk Extreme PRO 64GB UHS-I U3 170 MB/s and it performed worse when I ran benchmarks (using 240 MHz overclock on 700D):
Write speed:
1024 bytes: 82 MB/s
2048 bytes: 88 MB/s
4096 bytes: 89 MB/s
8192 bytes: 90 MB/s (matches 128 KB)
Read speed:
All was ~106 MB/s, no differences here.
Thanks Bilal. I don't know which bench.mo you're using and in which mode.
As for me, at 8192 bytes sector, recording speeds are the same as always: 58-59 Mb/s max sustained. I can't record in 2.8k mode at 10-bit.
I see it's the same Digic 5 chip your 700D and EOS M, and released about the same year (2012/13). Why are your benchmarks so high? 90 MB/s compared to max 76 MB/s with my 200 MB/s card, 70 MB/s in Liveview/Movie mode. Are there many differences between your build and Danne's that can affect card writing performance?
I'd like to know what recording speeds do people get with their cards, specially Extreme 170 MB/s, to know if my mine is underperforming.
Quote from: Danne on June 28, 2022, 04:08:04 PM
You don't get 72Mb/s in 192Mhz and 240Mhz don't work on eos m. Around 60 seems right.
If you do benchmark stay in photo mode and also push play button just before bench marking starts.
Quote from: theBilalFakhouri on June 28, 2022, 07:20:44 PM
Anyway, now your result matches 100D/700D with (Sandisk 170MB/s) at the same settings. RAW video recording should get up to 70 MB/s too (at least this happens on 100D/700D).
And again, your SD card seems fine.
I don't know if this an issue related to what things Danne have changed in his custom build which prevent getting 70 MB/s write speed, it could be his implementation/changes to small/new hacks (hacks don't work in some cases while recording RAW video?), or maybe other things?.
Why do Danne and Bilal implementations differ so much with the same Digic5 chip to yield a ~10 MB/s actual recording speed difference between M and 100/700D, when we get the same Bilal's-modified benchmark speeds? Default benchmark speeds also differ a ~10-13 MB/s. Can we expect to equal out?
@wyup
Try this, it should give you the extra 10 MB/s when using 192 MHz overclock:
crop_rec_4k_mlv_snd_raw_only_2022Jul31.EOSM202.zip (https://bitbucket.org/bilal_fakhouri/magic-lantern/downloads/crop_rec_4k_mlv_snd_raw_only_2022Jul31.EOSM202.zip)
Enable the hacks from RAW video submenu, then report back.
Share code please. Random builds without code is very bad practice.
It was online since 25/05/2022 . . check my implementation for the new hacks in 650D/700D source code (https://drive.google.com/file/d/1ZhlTQwCPHHb0QNqrpGHqfV8-NVqv30P7/view?usp=sharing), it's the same (you need to add EOS M addresses).
I don't keep the source code for these kind of tests, anyway I did this:
static REQUIRES(RawRecTask)
void hack_liveview_more()
{
/* Exlude Movie Crop Mode from the new hacks */
if (video_mode_crop)
{
return;
}
if (more_hacks) /* excludes mcm mode on eosm */
{
void (*aewbSuspend)() =
cam_eos_m ? 0xff2606f4 :
cam_5d3_113 ? 0xff23bc60 :
cam_5d3_123 ? 0xff23ff10 :
0;
void (*lvfaceEnd)() =
cam_eos_m ? 0xff177ff8 :
cam_5d3_113 ? 0xff16d77c :
cam_5d3_123 ? 0xff16e318 :
0;
lvfaceEnd();
if (more_hacks == 2)
{
aewbSuspend();
}
}
/* Causes freeze on eosm
if (one_more_hack)
{
void (*CartridgeCancel)() =
cam_eos_m ? 0xffa7e7d8 :
cam_5d3_113 ? 0xff17fd68 :
cam_5d3_123 ? 0xff181340 :
0;
CartridgeCancel();
msleep(10);
}
*/
}
Also reverted back "Small hacks" to official code to make sure all hacks would work for this test:
static REQUIRES(RawRecTask)
void hack_liveview(int unhack)
{
/* auto sets kill global draw on faster fps (EOSM) in real time preview mode */
if (kill_gd || (preview_mode == 1 && (shamem_read(0xC0F06804) == 0x2f701d4 || shamem_read(0xC0F06804) == 0x30701d4)))
{
if (!unhack)
{
idle_globaldraw_dis();
clrscr();
}
else
{
idle_globaldraw_en();
}
}
if (small_hacks)
{
/* disable canon graphics (gains a little speed) */
static int canon_gui_was_enabled;
if (!unhack)
{
canon_gui_was_enabled = !canon_gui_front_buffer_disabled();
canon_gui_disable_front_buffer();
}
else if (canon_gui_was_enabled)
{
canon_gui_enable_front_buffer(0);
canon_gui_was_enabled = 0;
}
/* disable auto exposure and auto white balance */
call("aewb_enableaewb", unhack ? 1 : 0); /* for new cameras */
call("lv_ae", unhack ? 1 : 0); /* for old cameras */
call("lv_wb", unhack ? 1 : 0);
/* change dialog refresh timer from 50ms to 8192ms */
uint32_t dialog_refresh_timer_addr = /* in StartDialogRefreshTimer */
cam_50d ? 0xffa84e00 :
cam_5d2 ? 0xffaac640 :
cam_5d3_113 ? 0xff4acda4 :
cam_5d3_123 ? 0xFF4B7648 :
cam_550d ? 0xFF2FE5E4 :
cam_600d ? 0xFF37AA18 :
cam_650d ? 0xFF527E38 :
cam_6d ? 0xFF52C684 :
cam_eos_m ? 0xFF539C1C :
cam_700d ? 0xFF52BB60 :
cam_7d ? 0xFF345788 :
cam_60d ? 0xff36fa3c :
cam_70d ? 0xFF558FF0 :
cam_100d ? 0xFF542580 :
cam_500d ? 0xFF2ABEF8 :
cam_1100d ? 0xFF373384 :
/* ... */
0;
uint32_t dialog_refresh_timer_orig_instr = 0xe3a00032; /* mov r0, #50 */
uint32_t dialog_refresh_timer_new_instr = 0xe3a00a02; /* change to mov r0, #8192 */
if (dialog_refresh_timer_addr)
{
if (!unhack) /* hack */
{
int err = patch_instruction(
dialog_refresh_timer_addr, dialog_refresh_timer_orig_instr, dialog_refresh_timer_new_instr,
"raw_rec: slow down Canon dialog refresh timer"
);
if (err)
{
NotifyBox(1000, "Hack error at %x:\nexpected %x, got %x", dialog_refresh_timer_addr, dialog_refresh_timer_orig_instr, *(volatile uint32_t*)dialog_refresh_timer_addr);
beep_custom(1000, 2000, 1);
}
}
else /* unhack */
{
unpatch_memory(dialog_refresh_timer_addr);
}
}
}
}
{
.name = "Small hacks",
.priv = &small_hacks,
.max = 1,
.choices = CHOICES("OFF", "ON"),
.help = "Slow down Canon GUI, disable auto exposure and af, white balance...",
.help2 = "Set af on to enable auto focus but loose som overhead.",
.advanced = 1,
},
Did this crop_rec_4k_mlv_snd_raw_only_2022Jul31.EOSM202.zip (https://bitbucket.org/bilal_fakhouri/magic-lantern/downloads/crop_rec_4k_mlv_snd_raw_only_2022Jul31.EOSM202.zip) work for you?
Did anyone get 70 MB/s write speed now on EOS M?
I think the problem in your current implementation for the new hacks in the following line:
if (more_hacks && shamem_read(0xc0f383d4) != 0x4f0010) /* excludes mcm mode on eosm */
But I am not sure, can't test, you should do some tests and figure out ;)
The implementation seems broken also in your 5D3 builds (tested UHD 1:1 preset and 1920x1280 @ 30 FPS at 14-bit uncompressed, the new hacks didn't work).
Only way for me to check against bugs is checking code. Any open published test build should include sources. Normally this change would be done in a branch derived from an original fork.
Not many players in here follow gpl basics which is a shame really. Code is much more interesting then the closed out fix.
Just run your code in git and all is good.
Quote from: Danne on August 01, 2022, 01:27:09 PM
Only way for me to check against bugs is checking code.
Hey, You did have the working code in first place (I PMed you my initial implementation back then when new hacks were discovered), my newer changes were included in my 650D/700D source code and it was online for months . . your changes broke the new hacks, have you tested your changes and made sure it's working before pushing your code? You should have noticed the problem, beside the working copy was online, you could compare it.
Quote from: Danne on August 01, 2022, 01:27:09 PM
Any open published test build should include sources. Normally this change would be done in a branch derived from an original fork.
Sorry, but I don't have forks for your repos, and I don't want to *support them, but I can report bugs and give some hints like what I did here.
*I am not a fan of your heavy ML modifications, most of them feels wrong, I don't want to support that (making commits) . . personally I think -to some point- you are following 1% fork style (https://www.magiclantern.fm/forum/index.php?board=58.0) , and it seems like a dead-end.
Notes from side (shared it on Discord (https://discord.com/channels/671072748985909258/671072748985909264/1002262209306570792)a few days ago, a user was asking for a merging possibility for your EOS M code to this reop (https://github.com/reticulatedpines/magiclantern_simplified/)):
QuoteDanne code for his EOS M build should be re-written from scratch, merging it at current state isn't a good idea, some notes from my side:
-Some of modules are loaded by default (you can't unload them), I see this as a downside
-Heavy ML UI changes (in particular Movie tab), and unfortunately to me he made it complex (not user friendly)
-crop_rec code needs major clean up, and again I think he made the code complex
-Several bugs which are easy to fix but not fixed yet
-Probably other things too
Quote from: Danne on August 01, 2022, 01:27:09 PM
Not many players in here follow gpl basics which is a shame really. Code is much more interesting then the closed out fix.
Again, my code is online, I didn't restrict the fix or any info.
Quote from: Danne on August 01, 2022, 01:27:09 PM
Just run your code in git and all is good.
I will run git, but as I mentioned, that does not mean I would work on your forks in my git repo.
If you prefer seeing the changes on git . . sorry, but that's not my problem, my code is online (on Google Drive, use SourceTree to see commits), does the GPL basics says I must use git?
Anyway, I said before I would rewrite my 650D/700D/100D code in a clean way and make the repo has native git support (no needs for hg / mercurial), and
actually re-writing in progress and it might take some time:
https://github.com/bilalfakhouri/magiclantern_hg_02/commits/crop_rec_4k_mlv_snd
My goal to include the latest achievements for all DIGIC 5 models in a clean way in one branch (includes 100D/650D/700D/5D3/EOS M and probably 6D and 70D). Beside a basic support for DIGIC 4 models.
Sorry if my reply did have a high tone.
You're missing the whole point. Strongly suggest you read up about gpl philosophy and workflow.
Couldn't care less about your support. Only want this. If you share builds or testing fixes out in the open to always share sources and clearly where to find it. That's it.
Also. No matter what you're opinions are you used a lot a lot of my work in your builds. Even if a lot of stuff is ugly code there's still the possibility to pick out the good stuff and build something new. That's how this works.
Okay, next time I wouldn't share test builds in public in these kind of cases.
But did you get ~70 MB/s write speed anyway?
Quote from: Danne on August 01, 2022, 03:14:27 PM
Also. No matter what you're opinions are you used a lot a lot of my work in your builds. Even if a lot of stuff is ugly code there's still the possibility to pick out the good stuff and build something new. That's how this works.
You didn't get the point, it's not about my "opinions", well:
-Due to your 5D3/EOS M builds, new users are getting more confused due to increased use complexity. I think this makes ML shape looks bad.
-How would contribute back to the main repo with your source code? Even if I decided to take the good stuff (to port EOS M to main repo), due to code complexity I might not able to do it.
I care about ML here.
Yeah there are a lot of shared code among us, you did that too . . that's about collaboration. If I saw some of
good code I will use it, and I will throw out the not good stuff.
Sorry but I did my best. It was even more confusing before.
As with any other piece of art you never finish it, you just leave it.
In this case it is fully possible for anybody to take my fork, create a branch, and start polishing whatever. My time is very limited for a long time to come :).
If a1ex or anyone with true coding abilities rather see that the fork is closed I probably consider that but so far this didn't happen. Personally I consider the code "safe" enough for using/playing/developing.
Quote
Anyway, I said before I would rewrite my 650D/700D/100D code in a clean way and make the repo has native git support (no needs for hg / mercurial), and actually re-writing in progress and it might take some time:
https://github.com/bilalfakhouri/magiclantern_hg_02/commits/crop_rec_4k_mlv_snd
My goal to include the latest achievements for all DIGIC 5 models in a clean way in one branch (includes 100D/650D/700D/5D3/EOS M and probably 6D and 70D). Beside a basic support for DIGIC 4 models.
My main ML repo has git support already. The commits that do that work are here:
https://github.com/reticulatedpines/magiclantern_simplified/commit/4a29a347f29276d4bec0876412cec4bb12b5a28c
https://github.com/reticulatedpines/magiclantern_simplified/commit/3163d41cee35fb54b4308e1f32318ff87de53645
(I think there might be a third commit near that time, too. But that should be most of it).
In theory, that repo should work on Digic 4 and 5 cams. It's been tested, a little... I'd be interested to learn if your modules work with it - but test in qemu first! You'll need to copy your module code to my repo and build there, because I made changes to how modules are compiled. The official repo has a serious bug that can make modules jump to incorrect address when doing calls, triggering bad crashes.
Do you know if crop_rec_4k_mlv_snd is very different from lua_fix? My repo has lua_fix already (and some other branches, merged together), so you may find it's not that hard to merge your changes into mine directly. You're welcome to take the hg removal commits whatever you choose. I think even better would be to recognise if the repo is hg *or* git, and work correctly for both. It's nice to have the commit tag in the build info screen, which mine currently doesn't do.
I don't think bilal is working from crop_rec_4k_mlv_snd branch.
Not sure if g3gg0 mlv_snd code is included into lua_fix but maybe it is. No time to check unfortunately.
Quote from: Danne on August 01, 2022, 04:19:44 PM
Sorry but I did my best. It was even more confusing before.
As with any other piece of art you never finish it, you just leave it.
In this case it is fully possible for anybody to take my fork, create a branch, and start polishing whatever. My time is very limited for a long time to come :).
If a1ex or anyone with true coding abilities rather see that the fork is closed I probably consider that but so far this didn't happen. Personally I consider the code "safe" enough for using/playing/developing.
Fair enough :)
For now, I am considering the Github repos as the way to go (Heptapod is dead currently), in particular:
magiclantern_simplified (https://github.com/reticulatedpines/magiclantern_simplified/) --> Currently used for porting ML to D678, beside D45 code works (needs more tests), this is based on unified branch, other branches have been removed to simplify things.
magiclantern_hg_02 (https://github.com/reticulatedpines/magiclantern_hg_02) --> This same as official ML mercurial repo (https://foss.heptapod.net/magic-lantern), but imported to git, it include all branches, in order to build these changes are required (https://github.com/reticulatedpines/magiclantern_simplified/commit/4a29a347f29276d4bec0876412cec4bb12b5a28c), I will be using a fork of it for now and will re-write the code there, and hopefully it would be included in either in here (https://github.com/reticulatedpines/magiclantern_simplified/) or here (https://github.com/reticulatedpines/magiclantern_hg_02).
Quote from: names_are_hard on August 01, 2022, 05:06:05 PM
My main ML repo has git support already.
Yes, I know, but it lacks crop_rec_4k_mlv_snd branch which is the one I am interested in.
This commit (https://github.com/reticulatedpines/magiclantern_simplified/commit/4a29a347f29276d4bec0876412cec4bb12b5a28c) was enough to let crop_rec_4k_mlv_snd builds without issues, I did it here (https://github.com/bilalfakhouri/magiclantern_hg_02/commit/d1f856ce6266b85c3188a2262361219fa35bdcf6), from initial tests everything seems to work without problems, I will follow your other commits for removing hg requirements.
Quote from: names_are_hard on August 01, 2022, 05:06:05 PM
In theory, that repo should work on Digic 4 and 5 cams. It's been tested, a little... I'd be interested to learn if your modules work with it - but test in qemu first! You'll need to copy your module code to my repo and build there, because I made changes to how modules are compiled. The official repo has a serious bug that can make modules jump to incorrect address when doing calls, triggering bad crashes.
Sure, I can do tests later, but I am not very interested in unified branch which is the main one you are using in magiclantern_simplified (https://github.com/reticulatedpines/magiclantern_simplified/).
Quote from: names_are_hard on August 01, 2022, 05:06:05 PM
Do you know if crop_rec_4k_mlv_snd is very different from lua_fix?
Yes, they are different, I don't how much but e.g. lua_fix doesn't have lossless compression or low bit-depths in uncompressed RAW, probably mlv_lite.mo doesn't work with mlv_snd.mo too.
Further more, adtg_gui.mo doesn't work in lua_fix (iirc OldAPI error?), but works in both crop_rec_4k_mlv_snd and iso-research branches.
I think lua_fix is similar to unified.
Quote from: names_are_hard on August 01, 2022, 05:06:05 PM
My repo has lua_fix already ..
Which repo exactly?
Quote from: names_are_hard on August 01, 2022, 05:06:05 PM
... so you may find it's not that hard to merge your changes into mine directly.
You mean to magiclantern_simplified (https://github.com/reticulatedpines/magiclantern_simplified/)? , I think we need to import crop_rec_4k_mlv_snd in magiclantern_simplified (https://github.com/reticulatedpines/magiclantern_simplified/) repo, then we need to find a way to merge dev branch and crop_rec_4k_mlv_snd correctly, but I think these two branches have a lot of differences, things might break, if that works I have no problem to work in your main branch.
For now I think the best way is to stay in my fork of magiclantern_hg_02 (https://github.com/bilalfakhouri/magiclantern_hg_02/commits/crop_rec_4k_mlv_snd), and port the new D45 crop mode stuff in one place, and probably merging other branches if it make sense, then importing the updated crop_rec_4k_mlv_snd to magiclantern_simplified (https://github.com/reticulatedpines/magiclantern_simplified/) then dealing with merging it to e.g. dev branch (unless if you have other plans).
Nice one Bilal 👍
Quote
Sure, I can do tests later, but I am not very interested in unified branch which is the main one you are using in magiclantern_simplified (https://github.com/reticulatedpines/magiclantern_simplified/).
Nope - unified is too old. Mine is unified + lua_fix + qemu + some other ones (I forget... there are so many open branches on official, I wanted to get away from that). I want one long-lived branch only (and stable / release branches, no work happens there). Other branches are for feature work only, before merging into dev (the main branch). I nuked a lot of history because it was too confusing to understand, hundreds of branches. I made that magiclantern_hg_02 repo to keep all history from all branches preserved in case it was needed.
Quote
I think lua_fix is similar to unified.
I don't really know, because I don't know of any documentation for what the different branches are for (other than commit messages), but in practice, lua_fix was the dev branch for several years. It has many changes that are unrelated to lua. I think the intention was to merge to unified, but this never happened.
Quote
I think we need to import crop_rec_4k_mlv_snd in magiclantern_simplified (https://github.com/reticulatedpines/magiclantern_simplified/) repo, then we need to find a way to merge dev branch and crop_rec_4k_mlv_snd correctly, but I think these two branches have a lot of differences, things might break, if that works I have no problem to work in your main branch.
I can look at this and see how different they are.
Quote
For now I think the best way is to stay in my fork of magiclantern_hg_02 (https://github.com/bilalfakhouri/magiclantern_hg_02/commits/crop_rec_4k_mlv_snd), and port the new D45 crop mode stuff in one place, and probably merging other branches if it make sense, then importing the updated crop_rec_4k_mlv_snd to magiclantern_simplified (https://github.com/reticulatedpines/magiclantern_simplified/) then dealing with merging it to e.g. dev branch (unless if you have other plans).
That would also work for me. I warn you that the different official branches are painful to merge. I had to inspect several 40k line diffs. Fun. Make sure you have some free time :D
crop_rec_4k_mlv_snd is very close to lua_fix, lua_fix was merged into it "only" 2 years ago:
https://foss.heptapod.net/magic-lantern/magic-lantern/-/network/branch%2Fcrop_rec_4k_mlv_snd
https://foss.heptapod.net/magic-lantern/magic-lantern/-/compare/branch%2Fcrop_rec_4k_mlv_snd...branch%2Flua_fix?from_project_id=321
Merging these would be easy, so merging to mine shouldn't be too hard; I've already done the hard lua_fix + unified merge.
But that's enough cluttering up this SD card thread! Let me know if you get the merge working, and I'll let you know if I do :)
Quote from: theBilalFakhouri on July 31, 2022, 06:12:19 AM
@wyup
Try this, it should give you the extra 10 MB/s when using 192 MHz overclock:
crop_rec_4k_mlv_snd_raw_only_2022Jul31.EOSM202.zip (https://bitbucket.org/bilal_fakhouri/magic-lantern/downloads/crop_rec_4k_mlv_snd_raw_only_2022Jul31.EOSM202.zip)
Enable the hacks from RAW video submenu, then report back.
I tested my own version and your hacks works just fine. ran 2.8k without hack 14bit lossless getting around 160 frames. Added the more hacks lvface + awb and getting continuous 2.8k 24 fps 14 bit lossless at around 69Mb/s so nothing wrong with my implementation. I suggest you test the 5d3 version a little more since I suspect it works just fine too.
Will add this to my code though. Much nicer than running shamem.
/* Exlude Movie Crop Mode from the new hacks */
if (video_mode_crop)
{
return;
}
Quote from: theBilalFakhouri on July 31, 2022, 06:12:19 AM
@wyup
Try this, it should give you the extra 10 MB/s when using 192 MHz overclock:
crop_rec_4k_mlv_snd_raw_only_2022Jul31.EOSM202.zip (https://bitbucket.org/bilal_fakhouri/magic-lantern/downloads/crop_rec_4k_mlv_snd_raw_only_2022Jul31.EOSM202.zip)
Enable the hacks from RAW video submenu, then report back.
I tried your build (overwrote files in card with yours) but I didn't notice improvement in actual recording speed. I get max around 60 MB/s with my 200MB/s Sandisk card.
You need to push existing presets as they are restricted to around 60Mb/s right now. Anyway, I already tested code and is working.
If I'm not mistaken, according to some reports people with Sandisk 170 MB/s get about 3MB/s faster than mine in both actual recording speed as benchmarks. So I'm not sure presets are restricted 60 MB/s, since as far as I know they get 63 MB/s.
Not restricted. Of you run the two hacks enabled with 2.8k preset and set bitdepth to OFF it will shoot into 68-69Mb/s.
I did. it doesn't make any difference. It ramps to 61-62 Mb/s briefly only to stop recording after 2-3-4 seconds.
Both hacks enabled, 2.8k Bit depth off.
I have the feeling you may want to purchase a Sandisk Extreme Pro R170/W90 to answer your questions.
AFAIK they come in all kind of sizes performing quite similiar and a 64 GByte specimen should do. Not like R200 variety which offers different write performance depending on size. Which one is yours, btw?
64GB one. According to specs is up to 90 MB/s write and 200 MB/s read speed. Write speed spec is the same in theory as the 170 MB/s one.
Is actual M crop_rec build tweaked to 'golden standard' 170MB/s card?
I actually don't know! What we know: There is no way to predict how any card type will react to ML's overclocking method.
So referring to any spec is not valid.
EDIT: I wouldn't be surprised if 256 GB, 400 GB, 512 GB, 1 TB varieties showed different results. They are specified R200/W140.
Yesterday I did some tests with Extreme Pro and Extreme Plus R170/W90. They perform quite the same and allowed transfer rates above 70 MByte/s. Didn't test "anamorphic" modes, though.
Samsung PRO Plus R160/W120 performed well in benchmark but really sucks trying to record with overclocking and higher transfer rates.
Quote from: wyup on August 04, 2022, 06:09:30 PM
I did. it doesn't make any difference. It ramps to 61-62 Mb/s briefly only to stop recording after 2-3-4 seconds.
Both hacks enabled, 2.8k Bit depth off.
at 2.8k, you won't get long clips at 14 bits. 10 bits will get you close to continuous and 12 bits will give you about 20-30 seconds. You can get more by underexposing the image and/or moving the camera slowly.
Actually, the latest hacks from thebilalfakhouri actually makes 2.8k continuous. Maybe even with 14 bits?
I tried last weekend, but I could not get it to record continuously under the sun (Wiith some shade, I got continuous at 10 bits) If you enable lvface+aewb, you're right, you'll get continuous at 12 bits, but you cannot use shutter fine tuning.
I reported the subtleties in this post: https://www.magiclantern.fm/forum/index.php?topic=26443.msg238437#msg238437
I will post some samples when I get time off, but when the exposure and focus are done right, 2.5k gives you reliability and plenty of satisfaction :)
p.s. I haven't benchmarked your latest (August 04 2022) build though. I did not see it in the commits, but did you add speed mods compared to the last build?
Quote from: theBilalFakhouri on August 01, 2022, 01:02:07 PM
Did this crop_rec_4k_mlv_snd_raw_only_2022Jul31.EOSM202.zip (https://bitbucket.org/bilal_fakhouri/magic-lantern/downloads/crop_rec_4k_mlv_snd_raw_only_2022Jul31.EOSM202.zip) work for you?
Did anyone get 70 MB/s write speed now on EOS M?
I'm a little late here, but here are my eosm bench results using the aforementioned build.
(https://i.ibb.co/CsN8tRT/bench1.jpg) (https://ibb.co/x8PLsxR)
With the lvface hack, I can get about 30 seconds at 12 bits, histogram showing 1.8 and even exposure (Sandisk Extreme pro 170 MB/s, 128 GB)
I know that not in the topic, but maybe someone will be interested. A few days ago I bought a SANDISK Extreme Pro SDXC 512 GB card on the Polish portal - SDSDXXY-512G-GN4IN for 60$ (279PLN). I tested it in a laptop reader and the transfers are the same as in SANDISK Extreme Pro SDXC 256 GB (purchased by me earlier in the official distribution). I also tested it in 5Dmk3 and with 240MHz settings in the record it has ~ 95 MB / s (analogous to the 256 GB version). The seller declares 14 days for a return and a two-year warranty (I assume that only in Poland, because we have such domestic regulations). I am not related to the seller in any way and I do not have any profit from this entry.
https://allegro.pl/oferta/karta-sdxc-sandisk-extreme-pro-512-gb-170mb-s-12543428724 (https://allegro.pl/oferta/karta-sdxc-sandisk-extreme-pro-512-gb-170mb-s-12543428724)
Hey, hey!
I know its an old topic... but I just now discovered this SD overclock thing... And I was wondering if someone tried it out with silent picture mode?
My aim would be to take more pictures... at the moment I think it needs 1-3 seconds to save the picture. So I can take picture only every some seconds with silent picture on. Is it possible to make more, like every second or this other mode "take pictures like crazy" ?
Thank to all of you beautiful people here! <3
Andre
PS : If anyway I can not take pictures faster, then probably I would buy a cheaper card. What standard would you recommend? UHS-1, V10, Class 10... sorry if this sound dumb.
First thing to do to accelerate silent pics: Write to MLV instead of DNG.
And you can use oc as well.
I just tried and got 18 FRSP frames in around 25 seconds.
Have you tried crop mood's "Full-Res LV" mode? 3 fps ...
I was using MLV. Accidentally I reseted my ML settings - took my some time to enable silent pic as I totally forgot how I activated it some time ago - and now the cam takes pictures every two seconds without stopping when I set it to 1 sec. Before it would stop, if there was not enough time to save the picture so I had to set the interval meter to 4 sec. So 0.5 FPS is already some improvement... but 18 in 25 would be better. :)
When I set the interval meter to "like crazy" it takes RAW pictures instead of MLV.
I searched for the OC modules on the cam... and found out I have only Nightly.2018.Jul03.EOSM202 build. What build would you recommend to use silent pic and oc mode together? I recently bought the cam as my old one was broken and this was the build that was there.
Maybe this is the reason I have no "Full-Res LV" in the silent pic menu. Only "Full-res", "Simple", "Burst"...
I could imagine using OC will make it only a little bit faster? Or? In this case I could stick to the old card instead buying a new faster one. Happy to hear opinion on this.
Thank you! :)
It seems to be you're the only M user never heard of this one: https://www.magiclantern.fm/forum/index.php?topic=26851.0 ;-)
"Full-Res LV" is a movie setting exclusive to Crop Mood build and not part of silent.mo.
You need a fast card compatible with 240 MHz oc frequency to take full advantage. My price/performance favorite is Samsung EVO Plus 256 GB (2021 edition, K suffix). Here is an list of tested cards: https://wiki.magiclantern.fm/cards_240mhz
Performance without oc: Around 38 MByte/ writing to card.
With oc @240 MHz: Around 84 MByte/s.
Those numbers cannot be achieved in silent mode, though. I get around 32 MByte/s on average.