Menu

Show posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Show posts Menu

Messages - KMikhail

#26
Quote from: dubzeebass on August 16, 2014, 03:30:33 PM
Except the 24-105 is regarded as a very very average L and isn't particularly sharp.

Indeed. There are tons of revealing tests, particularly the ones that compare it to a new 24-70/4, or even 24-70 2.8 II that I own. New lenses are head & shoulders above the 24-105 sharpness-wise.

The real bonus of sharp lens is the fact that in FullHD mode 5D3 isn't as sharp as in a 1:1 zoom mode and every bit of sharpness helps, while avoiding extra unsharp mask (which looks weird for FullHD, thanks to debayer - digital bayer) helps with having less noisy image too.

However, I do assume it might be hard to see the difference for not so contrasty image under imperfect conditions.
#27
As for the 5D3 RAW movie this cards looks impressive with 90+ mb write speed. Unfortunately it doesn't look like 29.97 FullHD is possible with this card.

Someone claimed a successful 30fps on 32GB SanDisk Extreme Pro. Would anyone try 29.97fps please?
#28
Got Lexar 1066x 128GB  - somewhat a disappointment.

In USB3/UDMA7 it showed stable 134 mb/s /write, but in playback on 5D3 (1.2.3) it was max 112mb/s, typically about 108. ~10mb/s slower than KB (117-119). Unlike KB card it never stalled in the beginning during recoding at 24fps, but couldn't sustain anything approaching 90mb/s (1:1 modes, 30fps, etc.).

I wonder if it is typical. Still waiting for my Lexar 64 GB 1066x  from Adorama. It seems like it's unrealistic to expect stable 90+ mb/s from 128GB cards?

#29
I own and use 85LII and 24-70L2.8II. The latter one is freakishly sharp in the center - moire is visible straight at 2.8. Not so good in the corners (at 2.8), but as zooms go - very, very good. So, as I stop down - center isn't really improving, neither sharpness nor micro contrast and this quality extends somewhere to sides.

24mm, f/2.8, 5D3, from RAW, obviously.
http://kartashovs.com/photo/2014_07_03_Randy_Tractors/2014_07_M03-2136.MP4

Once I push 85LII to f/1.8 it also becomes seriously sharp and contrasty in the center (actually this is my complain with respect to 85LII and 50L, they share this stopped-down behavior, I'd love to have it right at 1.2). At f/4 it is at optimum all over the frame and sharpness/contrast is top notch. Though, artistically it is not as fun anymore, I rarely go below 1.8. At 2.5 it is sharp everywhere, but not contrasty everywhere. Yup, I am the one geek who tests his lenses with MTF and TV charts! :) Bokeh is nice with both, with an edge to 85L. But in this case it is far, far less important than f1.2 vs. f2.8.

Hence, my point of view is: at current stage of tech zooms are very, very close to fixed lenses at the same apertures and at times actually even better (24L aint as sharp as my zoom at 2.8) due to necessity to be ultra-bright. So, it boils down to what aperture you need and how much money do you have. Though, out of newer 24/28/35 IS lenses only the last one is ~=  +- my zoom optically.

Regards,
Mikhail
#30
Hardware and Accessories / Re: Lexar x1066
July 17, 2014, 08:11:06 PM
Well, I didn't get mine (64GB) from Adorama. Got a couple of emails regarding back order situation, no more than this.
#31
Recently had a few glitches with the 64GB 95mb/s SD card from Sandisk as well. For example it couldn't finish writing the raw photo file and for the first time I witnessed an associated animation when tried to turn off the camera. It never finished, though, and I had to take the battery out.

Though, it occurred just recently, and might be associated with latest ML builds - hanged my camera a couple of  times.
#32
I ordered one of those and one Sandisk 95mb/s 64GB yesterday from Adorama. On their way to me!
#33
Has anyone managed to have fullhd raw recording at 30fps with 1.2.3 and sound enabled, including card spanning?
#34
Hardware and Accessories / Re: Lexar x1066
May 20, 2014, 11:16:53 PM
Any new owners of 1066x cards from Lexar and 5D3? Unfortunately only a few benchmarks are available and it would be nice to know their real-world speed.
#35
Quote from: Stedda on April 16, 2014, 01:45:29 PM
So much for reading ehhh?

Indeed, my dear reader. Plz read my statement again. I didn't not suggest to re-implement codecs, but merely the idea of how nice it would be to invoke what is already available. If it could be invoked from ML, of course. Canon FW routines can invoke it.
#36
Yes, it is normal for these KB 1066x 128GB cards and that's the cause of instability during recording at full hd raw. Cards I had never could continue recoding from the cold start, typically 83-85 mb/s would show up after a few attempts / bench runs. And then you need some luck. Well, not me - I returned them. But, I have to say and admit - that's still the best price/performance offering right now. Just +5..+10mb/s and it would be a winner for absolute majority of raw-video-shooters.
#37
Quote from: Sapphiron on May 05, 2014, 03:06:58 PM
Hello everyone!
Recently I bought KomputerBay 128GB 1066x CF and don't know if it's ok or broken...
Is this benchmark good or bad?

When GLOBAL DRAW is OFF you should get 115-120MB/s Write speed in playback, which would be in line with benches from KB and my own results.
#39
Hardware and Accessories / Re: Lexar x1066
May 02, 2014, 04:21:25 AM
Guys, pretty please, post your ML benches. 32-64-128 cards are most welcomed.

24fps movie and 30fps movie modes, preferably with Global Draw ON, 1.2.3 FW and, if by any chance, with SD spanning.

CrystalDiskMark and playback mode would be great too (W8 USB3 U7 readers).

Some of 1066x are sold at a discounted price right now, so it would super nice of you.

Thanks!

#40
Apparently:

a) Benchmark performance on 5D3 is somewhat different from actual recording performance (TS cards, barely around 80mb/s in rec mode, but can do 24fps).

b) 128 GB cards suffer in the department of speed vs. 64 GB cards and particularly with respect to consistency.

c) I've seen numbers for KB 128 1050x posted by KB on amazon as high as 134MB/s write and in my opinion that would be a true beast (toms hardware highest is 129mb/s for 64GB cards and they tried many readers too), but I didn't have anything approaching these numbers for KB 1066x or TS 1000x in a proper setting (w8 and U7, or 5D3 that can read in excess of 150mb/s).

I am debating if i should go for an expensive card. Obviously, I need to evaluate how much I like (and need) rawvideo, but will I get the performance I need from SanDisk 160mb/s and Lexar 1066x? Unfortunately there is no nice table of how many people have tested a specific card and kind of speed they got (people tend to report when they are strongly unsatisfied, seeking assistance, but not so often when they got what works for them).

I've read somewhere here that 30fps is possible with expensive cards and the question is - how dependable is it?
#41
Please let us know how it works! :)
#42
sometimes i could, sometimes i could not, regardless of warm up. same went for 1920*1152 24fps.
i'll order 64 cards to see how they do.
#43
Yes, 5d3.
#44
I tried 3x   128gb 1066x - all are about the same, pretty tightly.

Just installed 1.1.3 firmware and ML, it bumped up my writing rates by 2-3 mb/s (85-90mb/s, GD ON, movie). But 30fps 1080P with spanning still doesn't work (MLV).

Frames still get re-queued.
#45
Quote from: g3gg0 on April 16, 2014, 12:43:31 PM
its hard to make that UI-editable, but i can think of a config-file only option.

Thanks for your answers. I actually looked up for config already, fruitlessly, so it would be very nice to have!
#46
Ability to invoke raw lossless codec for raw video compression would be a heavenly gift: in 5D3 with 6fps it can pump through 22*6=132mpixels/s, while 1920*1080*24=49.8 mpixels/s, while reducing bandwidth requirement by 30-40%.
#47
A few comments on the already exceptional piece of programming mastery:

I have CF 128GB 1066x KB card and SD 128GB 600x Transcend (GD is on - peaking and histogram, only video MLV  mode considered)

a) 16384K buffer is generally slower than 15625K with my CF only (up to 88 MB/s, after several minutes of warm up).
b) In burnout mode some of the 25MB+ buffers achieved 90-91 MB/s in writing.
c) Going to 30fps eats a lot of bandwidth, easily 5-15MB/s for CF.
d) The fastest speed with SD I've seen was 18.5 MB/s, 20.1 MB/s reading.
e) Turning spanning on will degrade CF performance further and as a result SD card (a tad slowed too) doesn't help a whole lot.

Is there any way to select buffers for CF and SD for independent or spanning modes?
Sometimes I think having 1:2.40 would be nice, since 1920x800 is nicely divisible by 16 and actually used by some BDs.
I am still not entirely sure how to absolutely 100% guarantee video-to-sound sync. There is usually a difference in length between two tracks. And I strongly dislike this 24 vs. 23.976 thing, though I understand where it comes from.

At this point I'd say the only thing truly missing is ability to get 30fps with spanning, as I guess it is unreasonable to assume that any single card will sustain it along with a stable MLV with sound.
#48
The selling point?  :D
#49
Well, it is not good for me and I have contacted KB for a replacement. Though, their response was kind of strange - they didn't send me RMA# and form to fill out, just 'how we can help you'.

#50
In my Kingston USB 3.0 CrystalDiskMark gave me 136 and 111 mb/s for reading and writing, correspondingly.