Rendering Time

From Ultrafractal Wiki

also see:

   Networking
   Reducing Rendering Time
   Setting Number of Threads 
   with Slope and Lighting Formulas
   Slow PlugIns in reb.ucl
    

---

Hi Everybody,

I have a question, not related to the challenges.:

I am rendering a fractal, 15 layers, size : 2400x 1601 pixels, settings : anti-alias: custom treshold 0.5 depth  : 3 subdivisions 9 (3x3)

at this very moment it is still calculating... this are the figures: status calculating 0.07% elapsed : 0:17:27.53 remaining : 391:11:05.28 pixels/sec 2.73

is it really going to take more than 391 hours?? what am I doing wrong??

please..help.... i am so happy with this design ! thanx in advance juliette

***

To estimate rendering time:

For a good estimate, render the image at 1/10 scale: 800 x 615, with all of the settings and options the way you want them. Then, multiply that time by 100.

Kerry

***

Roughly speaking, increasing the Depth by 1 multiplies the time by a factor of 9; lowering the Threshold increases it even more. If you do find significant differences with the same UF4-compatible parameters and exactly the same settings. please post an example parameter set.

Kerry

***

"changed it to depth 1, and threshold 0.2.. it goes much quicker... will the Rendering Quality be less? juliette"

Hi Juliette

I see hardly any difference between quick and custom rendering myself. Render it twice the size I want to have. Resize and sharpen it with IrfanView.

Info in the UF pdf manual is on page 278.

Kind regards, Gerda

***

As mentioned, I think that your huge render times are based on two factors: your CPU might be old (Pentium 4s were first sold in 2000) and some of your layers are at very large magnifications. I'm not sure if you can do anything about the first issue (your CPU), but you can probably do something about the second (the large magnification).

For instance, your bottom layer (labeled Background) is rendered at a magnification of 4.4015926E15 (a very large #). But you can find a very similar pattern at a much lower magnification. For example, I spent about 10 minutes zooming out looking for a similar pattern and found one at a magnification of 4800000 (which is orders of magnitude smaller). You could probably find a similar pattern at an even lower magnification if you tried.

You could then try to find similar patterns for each of the other layers that have a very high magnification. This would reduce the render times dramatically.

FYI: On my machine your image rendered at 600x473 took 2 minutes 40 seconds. My machine is nowhere near the fastest (it is a dual core AMD 1.9 GHz). New machines can have CPUs with speeds above 3.0 GHz.

Here is your original formula with a second bottom layer (BG 2) at a reduced magnification (which you can compare to the original). It isn't an exact duplicate of the layer. It is just meant to be an example so that you could see that it is possible to find similar (or even exact) patterns at reduced magnifications:

pappasg@gmail.com

BromeliadNightsCroppedDuplicate {
::pcytDgn2tjVyOuNOQ07NQ/PI472N3XmAdYCQQwgZFYy9AaJabiWbhSuT7/+po2scs8E7eOMd
  A6gAE5isK9I57V8psxbSaMZ/093FF14aysxLevvM3m5MpR/hb7um6oEfZVlNNKdfVmLx0YXE
  9VXazuYBCFtzGmUMTSDlIzcw6rjVQO2UXTd8iPaL9btR/lpqyU/O6DiHIIk6d/cRq3ai+7GT
  yBIKvN6i7vrtAtoJ32srMNOffWjDSFghpqxVWEv43CzJisIqsyk4aOEjBckbhXTeZqNObfur
  wUkYv/uc4t6K22WvEbRj1HvEviIlUpUS0CmgwfYJaFSwFYBWJYSG8cUuZbRMMoeFhIwhsNFb
  h9mlYpYFVigQbK9A0MdQ18sLUbAHooKrPZnN5x4yNbi24ysFmcYXFWpFpGf6q9byXEBYxfIe
  xvDxsZr9lNLC1p6z1NGfTM6BoMfuq8rQNJtPv24yK33EjtE0935Kqdp2uzMvpoeDMviyCbUd
  ZmLNmRkKCmT0wUhkGnbqFyD2uIojplsft1XW2MmqmpFwWDfC0zeMf1+ksWI2DcfZtdJUjt2l
  ZuAhAgeiB2hWxhHbHojiI7WXJlZl+1QiZm6muNzQOmkYcY1lETD/qOGOKC/u9h+M37fy2f2B
  BfobGwOW3U26NpOY02FYmr4Rba8BbdUdOso21+IMXTLxBL1ROYD/5YugG1CpYCsXp0ox4iWq
  S7QYkSQQixc0k+B0YtQII9xJc1QtwItWyCVobIKhNULKmrEDhZDVKcA1xhbx/3H0DMeCnfEy
  Tje8lHw75TPg2zjGA6sFJA1TG461n0T1ntMuKjHOrCLwuKMIZLfy6hKPrgFtioJYlCDcaYnX
  xlCQjKoIBCzD0BMVSVwASKnxFggVRYEBhqldazgUeFDmsmI+AmPVKjZ85VykfMUy8jZV/F/Y
  SKQBToKysg+Uh8n8uR1YPMXdCM/A5WVZCcP5hJQD0fhkxoY9RiFWcUaopMkaIOoSGizRwyfQ
  aSooJSTRo/NekULJjaWJTqR0B6rcQbywaGWcN6tTg/ADGNB2zKUC4eG1GFNXUJ5CqNp6lq2w
  HVbc9/uYbXdmJN9Nt2Vq1GnLw7tmXkI79maXSQh1coyG/Lwiulrdb6KFrnlokzrcQ6JCkTFP
  Yg7NryBr5fjyZCbEjPV8ME/oEVIpCYb47rqmC+zFEA0vgqCmy4AM0EQPzdYAcnXVNRaiV4bR
  VxvkHzXbaIM/Hi7roHzy+c1LRK9n+1umP5NV1t6pwD7Mgoquy5NZQoUHkf7WS79YhZU6Thfz
  GcUGCtxb/ykZ0RnDb9um9p2+otq1kMwqbdTfq7826dlZphTZ+Qy9n7tbXm6KbSTfp7O5Gty2
  uc7eVFl3m4fZg+1ftkeQ9xpStAjmIXQDX4q5Aray1QD3PG+qHOd8OVOdaLDtQp1DCV4NOkkC
  L5Yy4Ayh2CYonAUCyRAsEP0ORCxFI23vzwprs56NgmXWHWUz5vdOTytLmZrSY1M385v0LiJ3
  gt3uik9/iv3elI0LpOG6JO+rWIMnO85NjNP+V4L9cpl1JlVdf63h8ONmakqfI3b3kFkC9/r7
  J74QzvCh/iJYAwkzWeMRYxzJKGX1/ZhtlIH+wBQLdqU7NPGvYPGY8wX1yw0L7dfUl/NOQoyL
  Y/QRvs3dC9Ce3p6b27+Jw/oxB1EgPvPD6s2SU0ZcfEA88+Mo6XaTD2l8ZAn0PaLeznxrGfGr
  Lf+/mJDHsx7KM+DvqNZwHvjnwZChgNnJDMC0+U5oSGpGTSBOTU8ZtZgRKw5ygQdpeMJa4oRS
  OzmhOkCTM1lxwQCCgGp6mcZwx3kLDk66dZoV3mJD8t3v49fcqDD4F+kr2tOz2eYHdudjL24Q
  zIUGl03ogRlwZWfTBFK8n++BAB4HyWBvWuSdkpP4mmJhm2sh/vglS8RtxVcF3kydOjSOlnhI
  XJjyk84Wf5+i0T4VXPT6trgej39N8u/BskywCC==
}