More details please! I’d like to make those tweaks
I use the negative film simulation (or the positive one)
I thought I tweaked jpeg contrast but I guess not.
Thanks
In Fuji land, I would recommend getting a 24 MP (X-Trans) body at least. They were a really significant jump from the 16 MP ones. Not just on resolution, but a good stop less noise at high ISO, much faster focusing, noticeably less shutter blackout. And most of them came with USB charging (all except X-Pro2, IIRC), which is positively indispensable for me personally.
3 Likes
I think I’ll save up my hard-earned cash and either get an X-T30 or an X-T5 (or whatever the latest versions are by the time I’m finished saving). It looks like the improvements over my aging X-T10 would be well worth the investment.
And, if I work REALLY hard (and stop spending so much money on junk food and other such rubbish), I may even be able to get a used GR as well! Or maybe I’ll just run a ‘Kick up the Starter’ campaign: “Hard-up photographer needs a new camera — send in your pledges and you’ll get a free photo (of their happy little face!) to hang on your wall!” 
2 Likes
I’ve just bought a new camera… for $50 AUD! A Pentax Z10 SLR… looking forward using a few more rolls of HP5+ 
Hoping it really is in working order…
1 Like
So you want to buy a new camera?
No. Sony A7Rii (end hand) here and pretty happy with it. This said, the A7Riii (2nd hand) is now below $1000 in my area and tempting.
I feel like it’s more the display resolution (and thus the preview size) that impact Darktable responsiveness.
I upgraded to a 4k display a few month ago and Dt performances took a serious hit … Now I always use the “ISOxxx assessment mode” to reduce the preview area.
I did use it before to asses the contrast/white ior rather max and overhaul brightness of my picture perceptually but just toggled it to check from time to time, now it became a performance tool as well 
PS : otherwise, the added real-estate I gat upgrading from 1920x1080 is really great as I can see so much more interface and tools and so much more picture as well when I want to pixel peak to fine tune micro contrast/sharpness
Did you upgrade your graphics card at the same time?
Have fun!
Claes in Lund, Sweden
Yes, I’ve read similar things. My screen is 1440p, so that helps, along with the RTX 2060 Super card.
Also, if you shoot video, the differences between old and new models tend to be more significant.
The primary reason I upgraded from an A7M3 to an A7M4 was 10-bit video - there was nothing in the stills functionality I was particularly interested in. The more minor (but REALLY nice to have) thing was that the A7M3 had USB-C but did NOT support USB PD - and in fact would only consume around 0.8A at 5v when the power switch was in the “On” position - it would routinely drain the battery on external power. A7M4 supports USB PD which is a huge improvement.
Funny thing, most of my purchases lately have been cleaning equipment… No pets, but after 15 years my old car (which I’ve kept as a backup/beater) really needs some serious cleaning. I recently purchased a carpet shampooer (Bissell Proheat Revolution) and a steam cleaner (McCullouch MC1375) to take better care of my cars.
Which reminds me, I need to stop procrastinating on adding generic-brand Dynamat to my new car to dampen the road noise.
I did not though that my gtx 1650 would not be up to the task …
What would be the requirement to have average perfs on Dt with a 4K display ? How to diagnose of it’s not my CPU that is the bottleneck ?
I may change for an a2000 if I manage to find a good deal … I don’t want a GPU with extra power plug as I have a very low power PSU and intend to keep it that way 
My GPU is clearly not in the same leage by at least an order of magnitude
Well, maybe my card is grossly overspecced. It’s the only one I’ve had, and came with the secondhand gaming rig I bought from a local lad who was leaving for college. It also came with some bare chested selfies of him flexing his biceps on the desktop as he failed to wipe the drive…
1 Like
Nice catch … for the GPU.
1 Like
Here’s something to chew on. If you don’t want to buy a new camera, you might alternatively think about what processing you need to make the old camera palatable. In that regard, older noisier cameras need decent denoise, and the good algorithms for that operation are really compute-intensive. My limited experience coding these intuits that CPU-based denoise can be made digestible (> 30sec), but GPU-based alternatives can cut that by an order of magnitude. If the majority of your old-camera images need denoise, you might at least be buying a GPU and selecting software that uses it.
My old D7000 would put me in that bucket. My “new” Z 6 not so much; I think I’ve only had one image from it so far that suggested denoise, and it was a salvage operation for a mis-calculated exposure. Otherwise, it’s pretty clean in the shadows, keeping my feeble hack software in business…
Yes, this a veiled PSA for vkdt. It only makes sense, load the image in the GPU from the gitgo, only pull it out for export after all the ops are done as GPU shaders. Well-worth the $300US investment for the card. The CAD software I use is on the cusp of adopting a GPU-based mesh library, which will knock-down my longish render times significantly.
Sorry for the somewhat-wandersome missive…
5 Likes
Lolz. I decided against investigating the rest of the drive before wiping it
Let’s remove the veil.
vkdt is very, very good.
I have a relatively budget-level GPU
NVIDIA Corporation TU116 [GeForce GTX 1660] (rev a1)
and on a 4k monitor (3840x2160) vkdt is instantaneous on most operations
Also, I’m too invested in bodies and lenses to change to mirrorless.
1 Like
… i really hope on your machine vkdt doesn’t take a whole second to denoise images! on a 24MP image i get:
[perf] sum denoise: 2.985 ms
so that’s a couple orders of magnitude.
1 Like