SoylentNews is people
SoylentNews
NVIDIA Announces the GeForce RTX 20 Series: RTX 2080 Ti & 2080 on Sept. 20th, RTX 2070 in October
NVIDIA's Gamescom 2018 keynote just wrapped up, and as many have been expecting since it was announced last month, NVIDIA is getting ready to launch their next generation of GeForce hardware. Announced at the event and going on sale starting September 20th is NVIDIA's GeForce RTX 20 series, which is succeeding the current Pascal-powered GeForce GTX 10 series. Based on NVIDIA's new Turing GPU architecture and built on TSMC's 12nm "FFN" process, NVIDIA has lofty goals, looking to drive an entire paradigm shift in how games are rendered and how PC video cards are evaluated. CEO Jensen Huang has called Turing NVIDIA's most important GPU architecture since 2006's Tesla GPU architecture (G80 GPU), and from a features standpoint it's clear that he's not overstating matters.
[...] So what does Turing bring to the table? The marquee feature across the board is hybrid rendering, which combines ray tracing with traditional rasterization to exploit the strengths of both technologies. This announcement is essentially a continuation of NVIDIA's RTX announcement from earlier this year, so if you thought that announcement was a little sparse, well then here is the rest of the story.
The big change here is that NVIDIA is going to be including even more ray tracing hardware with Turing in order to offer faster and more efficient hardware ray tracing acceleration. New to the Turing architecture is what NVIDIA is calling an RT core, the underpinnings of which we aren't fully informed on at this time, but serve as dedicated ray tracing processors. These processor blocks accelerate both ray-triangle intersection checks and bounding volume hierarchy (BVH) manipulation, the latter being a very popular data structure for storing objects for ray tracing.
NVIDIA is stating that the fastest GeForce RTX part can cast 10 Billion (Giga) rays per second, which compared to the unaccelerated Pascal is a 25x improvement in ray tracing performance.
Nvidia has confirmed that the machine learning capabilities (tensor cores) of the GPU will used to smooth out problems with ray-tracing. Real-time AI denoising (4m17s) will be used to reduce the amount of samples per pixel needed to achieve photorealism.
Previously: Microsoft Announces Directx 12 Raytracing API
Nvidia Announces Turing Architecture With Focus on Ray-Tracing and Lower-Precision Operations
Related: Real-time Ray-tracing at GDC 2014
(Score: 4, Interesting) by MichaelDavidCrawford on Tuesday August 21 2018, @12:45PM
Also complex reflectors such as the Ritchey-Chrétien [starizona.com] of which the Hubble is an example. Most big scopes other than the Palomar 200" are Ritchey-Chrétiens.
The main advantage of the Ritchey-Chrétian are a long focal length - high magnification - with a shorter tube than is possible with the Cassegrain [starizona.com], as well as a much wider field of view. The disadvantage of the Ritchey-Chrétien is that it can _only_ be used with that one long focal length, whereas with the Cassegrain one can remove the Hyperboloid secondary mirror then have a flat focal plane for a much shorter focal length but with a poor field of view, or possibly with a 45-degree mirror for a Newtonian [starizona.com], which was the most common type of commercially available scope until the early eighties.
The very _first_ C program I ever attempted to write was for lens ray-tracing.
These kinds of programs generally include automatic design optimization. In principle you can give it the Schott Optical Glass catalog, tell it what you want in the way of a focal length, field of view and how many lenses you want - a scope for eyeball astronomy or a spotting scope has two elements, a biconvex Crown Glass lens and a concave/flat Flint Glass lens for reasonably good correction of Chromatic Aberration whereas the Cook Triplet [briansmith.com] is commonly used for wide field of view astrophotography.
As you might expect I didn't do much ahead of time design on that first ray tracer however I did think about it a lot starting when I was thirteen. I took Physics 20 Computational Physics then on a 16-Bit 8086 IBM XT I wrote the entire program before I tried to get it to compile. The first time I tried to compile there were oodles of errors. It took me about a week to fix all those, then I tried to run it and of course got a GPF.
At that point I gave up. Fortunately I did a whole lot better with Pascal that same term in CD 10, Introduction To Algorithms in which our final project was a full-featured color vector graphic editor that didn't support saving to disk or printing but was otherwise equivalent to the 8-Bit color MacDraw, but with a tablet interface.
We used the 68000-based HP Chipmunk with the UCSD P-System. The 8086 DOS XT totally 5ux0r3d compared to the 68k with the P-System. And in fact I still have all my floppies for both boxen and will be getting them out of storage sometime later this year.
I used to have a real extension page about amateur telescope making but then that site dropped dead. I'll put up a placeholder page [warplife.com] then move my old pages over to warplife.com this weekend.
When I lived in Owl's Head, Maine - in the Midcoast region, near Rockland - one of the few things that gave me real relief from the Dot-Com Crash was to work on an eight-inch Ritchey-Chrétien that was intended to be a scope I could take in airline carry-on luggage. That's a real good application for an amateur RC.
But I've decided to abandon the effort then start all over again in part because I drilled too far through the primary so that the plug fell out. You don't want to do that rather what you do is to drill most of the way towards the end of fine grinding so as to relieve stresses in the glass then cut all the way through after you're done figuring. To cut all the way through is not insurmountable as you can glue the plug back in with beeswax but it can be troublesome due to edge effects around the center hole that are quite difficult to correct during polishing and figuring.
The other reason is that I found out only after starting work on my 8" RC that one can take ten-inch scopes on a plane. The ratio of luminosity scales as the square of the diameter in this case 100/64 or 1.563. In my actual experience of having built both 8" and 10" Newtonians is that the extra effort required for the larger mirror is well worth it.
DO NOT ATTEMPT AN RC FOR YOUR FIRST SCOPE! YOU HAVE BEEN WARNED.
Don't attempt a Cassegrain either. It is a huge PITA to test RC and Cass secondary mirrors and so are quite advanced projects. Make an 8" F/8 - that is, a 64" or five foot four-inch focal length Newtonian.
Such a long tube won't fit in most of today's cars and so is mostly suitable only for backyard use. Make a short ten inch for your second scope. You really don't want really low F-Ratios for the larger mirrors - the aberrations don't scale the right way. Really the shortest you should make a ten-inch would be 50" or 4" 2" which would fit in most cars either if you have a wagon, a seat that folds down to expose the trunk or a cargo bin on the roof.
You want the long F-Ratio for your first scope because the final figured curve must be a Paraboloid Of Revolution. The Foucault Test works really really well for figuring spheres but again is a huge PITA for any other kinds of curves. But the difference between a sphere and the required Paraboloid is smaller and smaller as the F-Ratio increases. It was once quite common for beginners to start with a 6" mirror with a 96" focal length, in which case there is no discernible difference between a sphere and an optically-perfect paraboloid, but again in my own experience the 6" doesn't capture nearly enough light and so is quite disappointing for nebulae.
The 6" works well for Venus, Jupitar and Saturn but not for Mars. For Mars you really want a 12" or larger mirror and a long focal length with a clock-driven equatorial mount.
Mirror grinding kits do NOT come with instructions. You need a book [willbell.com]. I recommend you purchase Jean Texereau's How To Make A Telescope [willbell.com] 2nd Edition.
The 1st worked out real well for me. I used Texereau's recipe for chemically silvering my 6" mirror but again silvering is a huge PITA, it requires cleaning the glass with Fuming Nitric Acid which would get the FBI breaking down your door if you don't obtain it the right way and the very slightest trace of impurities will totally ruin your coat.
Instead get your mirror Vacuum Aluminized for $50 or $75 dollars. There's lots of shops that do that, many of them operated by other amateurs. (Aluminizing again is an advanced project).
I was supposed to be recording Michael David Crawford LIVE! On Broadway [soggywizards.com] (and Morrison, Portland, Oregon) The Rough Draft right now. It's working well so far so I'll stop writing now.
Yes I Have No Bananas. [gofundme.com]