Any plans to fix RAYTRACED mode or is it now where its gonna stay ?
Raytraced handles transparency a bit different (for i.e. glass and transparent plastic materials).
The transparency color should be regarded as a filter, where white lets all light through. This means that black lets no light through, no matter how high you set the transparency.
Since the Rhino color shows RGB values in the range of 0-255 per channel we’ll use that for the rest too - internally colors are mapped to 0.0-1.0 where Raytraced actually can handle values outside the range (for high dynamic range)
Now, if you want a black glass with 50% transparency you will need a value in the middle between black and white, that is RGB (128,128,128). That as your transparency color with transparency set to 100% gives you the black at 50% transparency.
If you want black 25% you set transparency to 100% for a color value of (64,64,64).
For a dark grey at 25% the maths will be similar, but you have some play room by setting transparency to something lower than 100%.
This all is a result of more physically correct rendering, and it will be the direction materials are going to for the future, at least for Raytraced, and when eventually the underlying engine replaces to trusty old Rhino Render.
THANKS for your reply. I GENUINELY appreciate the GOOD work done by RHINO developers !
BURIED in all of your JARGON (look it up) BASIC RHINO MATERIALS no longer work as INTENDED ? NEON had NONE of these issues, but instead, WITHOUT waiting THREE to FIVE MINUTES between_zooms gave a FAITHFUL representation of the RENDERED Viewport with the attendant INTENDED increase in fidelity. What we wait THREE to FIVE MINUTES for with an UNUSABLE RESULT we USED to get in less than 30 seconds (still DO in -5), and we could PROPERLY use “Rendered” viewport as a REAL-TIME PREVIEW of what we get when we reset the viewport to NEON. NOW we WAIT and wait and wait and … for an unusable result that we can NOT preview in the Rendered Viewport, and can NOT control with basic NO WEIGHT Rhino material settings ? As a “relacement” for NEON as a “Real-Time-Renderer” RAYTRACED is a complete FAIL. As a replacement for RHINO RENDER, if it will NOT adequately handle the view as previewed in Rendered Viewport, while for SOME applications it may be superior, NOT HANDLING BASIC MATERIALS correctly- its LOSS of PREDICTABILITY makes it a genuine PROBLEM that can only be solved by PURCHASING a commercial Renderer for MULTIPLE HUNDREDS OF DOLLARS, AGAIN a FAIL.
Consider - Someone made the CONSCIOUS DECISION to make this change. Pause and let that sink in for a moment. NOW consider - the ONLY likely response to THIS reply will be an eyes-glazed-over ignore.
Finally consider - It DID NOT escape my notice that the actual question posed was dodged in favor of a JARGON-FILLED response that offered no solution to the problem initially reported. NO ? So WHAT would you ACTUALLY DO to give transparency to the parts THAT ARE NOT “BLACK” (???) but rather HAD_a_transparency at 2 full minutes AFTER the render reported “complete” ? COLOR controls TRANSPARENCY LEVEL ? Who’s idea was THAT ?!?
The FRUSTRATING THING that actually prompted my initial communication was that the FIRST result DID show TRANSPARENCY which was BEAUTIFUL (and COULD double as a replacement for RHINO RENDER) but the re-run NEVER ACHIEVED TRANSPARENCY, just as reported. Fail.
NO response to the actual question. No suggestions as to ACTUAL solution, JUST a JARGON-FILLED rationalization for someone’s personal direction change decision.
BASIC materials support with the change in approach described above an available optional method ? NO ? Why ?
AND it still takes WAY_TOO_LONG anyway, with “promises” to make THAT better … in the future… (RTR ?)
Look, you can (and probably will) dismiss this as well, “WE are the DEVELOPERS so WE are smarter …” . Well, my only response is that- I am a 14+ year RHINO USER (since the early days of RH-3.0) and have RUN CAD groups on MULTI-MILLION Dollar Projects for LARGE companies, and have dealt with this phenomenon my entire career, STARTING with the EARLY DAYS of AutoDesk, and the phenomenon is as long-standing, CONSISTENT and entirely as PREDICTABLE as it is FRUSTRATING.
THIS could have TOO if you’d have listened to the feedback from ALL of us who’ve tried raising these concerns, and it still could … as an EFFECTIVE replacement for RHINO RENDER. AS a replacement as RHINO’s Real-Time-Renderer, it still falls short for the reasons included here and in ALL of the posts on the subject from myself and other users posting their questions and concerns in Serengeti.
DNR-TL ? Kinda proves my point ?
Charles S. Fee - 281-546-5091
3D/2D CAD Designer,Modeler and Animator
NEW Product Development
PatentSupport a Specialty !
What graphics card do you have?
Haven’t tried neon…
when is support of neon end? Just curious…
We will not be releasing a version of Neon for Rhino 6
Hmm… I guess one would feel left in the dark a bit if been using a better renderer than rhino render during V5…
if cycles can be put in V6 as WIP plugin that’ll be nice. tho I suppose integrating and plugin would be a different development…
Raytraced (Cycles) is in V6.
right. sorry about the misleading comment. my bad…
was thinking neon was a full render plugin, with material view and render window.
NEON is a FREE NRQ-RTR (Near-Render-Quality - Real-Time-Renderer) PLUGIN for the RHINO Viewport.
Download it as a FREE PLUGIN from RHINO. What it DOES is step your viewport up one level from basic “Render” to an improved image that looks very much like the rendered viewport but with near-RENDERED quality.
NEON is ONLY available for RHINO - 4 & 5 , NOT for RH-6. “Raytraced” is the planned replacement .
I find NEON EXTREMELY useful as a manageable improvement over basic “Rendered” Viewport for creating still images WORTH CAPTURING for publishing AND as an EXCELLENT display for RENDERING BONGO ANIMATIONS. The HUGE advantage of rendering an animation in - STANDARD_Rendered_viewport_ is that you get EXCELLENT image QUALITY in your animation, and if THAT image is good enough for your purposes, you’re ALL SET at 0.0 (ZERO POINT ZERO) seconds per frame ! IF that image is good, but you need something just a bit better (improved material representation, lighting, shadows, etc.) with only a BIT of a time hit (about 30 SECONDS per frame) then NEON is a WONDERFUL solution. A FULL RENDERER at 3 - 5 minutes OR MORE per frame when NEON gives you (seriously) NEARLY the same level of quality at 1/5 to 1/10 the time hit, NEON becomes a real contender. ADD to that (for comparison purposes) 5 + minutes between keystrokes frame-to-frame for the NEW and IMPROVED “Ray-traced” replacement for “NEON”, with BAD “Noise” RIGHT UP TO THE END of the render with NEON giving a useable image at 15 seconds, a nearly COMPLETE IMAGE at 30secs and a FULLY FINISHED image at just over 1m:30sec) . STANDARD RHINO-Material TRANSPARENCY doesn’t seem to be supported in “Raytraced” but its PERFECTLY supported in Rhino STANDARD “Rendered Viewport” AND NEON. Not Raytraced, INCLUDING Incidence of Refraction.
ADD the FACT that the WONDERFUL folks at McNeel made this plugin available to you for FREE (as opposed to $350, $750, $1500, $2500, $3500 + + + for a commercial renderer ) makes it a REAL contender ! Now, REALISTICALLY its prob’ly NOT suitable for COMMERCIAL production (MAYE) but I have REPEATEDLY found it to be the IDEAL SOLUTION for EVERY SINGLE INDUSTRIAL ANIMATION I have produced, with STANDARD RHINO “Rendered Viewport” a CLOSE CAPABLE SECOND !
PLEASE don’t misunderstand - Raytrace holds HUGE potential, and I genuinely look forward to its eventual success !
I hope this helps, and as it ONLY represents MY implementation on a CAPABLE but these days a generally mid-stream computer - it MAY NOT equal YOUR experience, as I HAVE NOT yet upgraded my video to the new generally preferred $1000+ solution that may be the real solution to the dillemma.
Best of luck to you and others on this journey !
It’s not free. We’re going to charge you for this stuff.
You are going to start charging for Neon - 4,5 years after its last update?
I kind of had the same feelings about Raytraced versus Neon. Then I found that my GPU - a AMD FirePro V5900 (FireGL V) – is far too old (2009) and drivers no longer supported by ADM. Hence a very low performance of ‘Raytraced’. That’s why Andy asked you about your GPU.
Also currently the OpenCL performance for unknown reason is far below par, even on strong, recent AMD hardware (I am investigating to find the cause of that - Cycles in Blender on my AMD RX 550 runs just fine, so in theory it should in Rhino as well).
Does your CPU rendering give better performance? I just realized that your card is pre-GCN2 anyway, so it won’t even show in the OpenCL tab for device selection. This means you are by definition rendering on the CPU.
So I guess the best thing I can do is invest in a decent GPU.
Any sugestions - apart from the AMD RX 550?
If you already own one of the newer AMD models, I would just wait a bit until @nathanletwory has figured out what holds them back. like he mentioned, in blender/cycles they deliver comparable results to same gen nvidvia GPUs.
if you want to upgrade anyway (because @Luc card is WAY TOO OLD), and don’t want to wait for this to be fixed, then I would recommend any Pascal based nvidia GPU from GTX1060-6GB (don’t take the 3GB version since it not only has less VRAM but also less shaders) and up. Just depends on how much you wanna spend.
If you want to spend more, then you have (roughly) the following options:
keep in mind, that spending a little more on a good custom card will most likely result in better thermals and an overall quieter system.
Btw, what’s also not so nice about AMD cards at the moment is that they are still quite overpriced because of the current mining boom. But this only affects RX470, RX480, RX570, RX580. Lower end cards like RX550 and RX560 are not affected.
I’m quite happy with my GTX1060 at the moment. The Zotac-mini model which I chose is very quiet and the GPU is overall very energy efficient.
I consider switching back to AMD, maybe a VEGA, but only if there will be a “Nano” version available like the R9-Nano. My case is just too tiny for big cards.
And of course I’ll wait until nathan has figured out what’s holding back the AMD GPUs.
Here are some benchmarks:
the bar graph is nearly impossible to read. just click all the GPU and CPU models you don’t want to compare - they’ll be crossed out and the graph gets more readable.
Just a comparison page 1070ti or 1080 would be quite fast.
1060 or 1070 seems average…
1050 if monitor is like 4K would be a bit a little slow
Heh well if you wanted to really see what Neon could do you needed to own Brazil. And a Caustic board, those were cool. Too bad the whole ‘professional market’ thing was just a test bed/promo for tech Imagination wants to put in our phones…well and it was probably inevitable they’d lose out to CUDA solutions.
As opposed to ALL previous replies to my queries proclaiming its NO LONGER AVAILABLE and that THERE ARE NO PLANS to make it available, NOW it suddenly WILL be available AND you’re going to CHARGE FOR IT ?
NO further plans to correct the MOUNTAINS of problems with its replacement, but now plans to CHARGE for a plugin that has not seen further improvement since it was offered as an available plugin for RH-5 ?
Look, I’d be willing to PAY a reasonable price for this, as its replacement (Raytraced) is largely a no-start FAIL, but SERIOUSLY - take a deep breath, PAUSE for a moment and SERIOUSLY - TAKE A CLOSE LOOK_ at what this MUST MEAN about your decision making process to your USER BASE ! Interestingly, MY reaction seems NOT to be the first one posted. Do you SEE that you have just proven in ONE (probably inadvertent) statement that EVERY SINGLE CONCERN I state is fully and properly ACCURATE ?
Interesting to ALSO note that while you admit plans to CHARGE for this, you DON’T mention a pricing level ? Happy to pay a reasonable price but even THAT won’t change what’s gone before.
I’d be even MUCH more eager to pay a REASONABLE PRICE for something that was previously offered for free IF the offer to SELL was from the ORIGINAL DEVELOPER who is looking to sell it outside the McNeel enterprise (because someone made the internal decision NOT to pay the developer what he was asking for it justifying a full-price re-sell thru the McNeel portal as a tested and approved Plugin), but would be disappointed that my personally favorite Software developer wouldn’t be able to benefit, ESPECIALLY as a result of what seems like a speed-bump marketing decision.
I’m hoping this doesn’t get posted to Serengeti, as this is a response to a direct e-mail, and my complete intent is to offer helpful input. NO insult WHATSOEVER is intended. I am a MANY YEARS (15+) user AND FAN of RHINO, and SINCERELY hope for McNeel’s continued success.
December 1 |
Heh well if you wanted to really see what Neon could do you needed to own Brazil. And a Caustic board, those were cool. Too bad the whole ‘professional market’ thing was just a test bed/promo for tech Imagination wants to put in our phones…well and it was probably inevitable they’d lose out to CUDA solutions. Visit Topic or reply to this email to respond.
In Reply To
December 1 |
You are going to start charging for Neon - 4,5 years after its last update? Visit Topic or reply to this email to respond. To unsubscribe from these emails, click here.
Uh I think there is a bit of misunderstanding here…