subb

On a single core processor, obviously not, but massive C++ codebase are compiled with multicore CPU, and often distributed. The advantage of the header inclusion model is you get an bunch of independent compilation units, and so it's straight forward to parallelize.

A common way to speed things up even more is to merge some cpp file together, so each compiler thread is active for longer and common headers are compiled once.

With the new module feature of C++20, compilation is transformed into a DAG, and it is still unclear if it's really faster than just throwing a lot of cores at the problem.

Isn't this pretty much the same as taking a loan at a bank and buying bitcoins? Couldn't the "bubble" be explained by extremely low interest rates?

Banks create money out of thin air when you take a loan. The only difference is they destroy that newly created money when you pay the loan back (but keep the interests).

See : https://positivemoney.org/how-money-works/banking-101-video-...

Absolutely. At some point, the electrons in the transistor experience quantum tunneling - i.e. they jump over the gate. Which means it's not a very useful gate anymore...

See this great video about it : https://www.youtube.com/watch?v=67S4IyakRko

Not much evidence, but they stopped reading emails to target ads a while ago. https://www.bloomberg.com/news/articles/2017-06-23/google-wi...

https://support.google.com/mail/answer/6603?hl=en

Why not Arq with their own cloud plan? What's the advantage of using their tool with another cloud storage provider?

Well, he started it.

I wonder if they could take the mesure from the underside of the plane (90 degrees difference from the front)? It might help for bird strike, not much for ice.

Indeed, that could do it I think.

I'm not too knowledgeable about patents, but maybe we need a "liberal" patent? Something that can't be used to sue other people, but protects you from being sued.

Np. All this marketing show is quite elaborate and it's hard to get true answers. Thanks for that top level comment by the way, this whole thread is crazy with speculation and misunderstanding.

In DXR, you can specify a custom intersection shader, thus it's not only limited to triangles.

It's a strange message that one. On one hand, we can totally do raytracing now without any specialized API. Screen space reflection is raytracing, but only on what's visible on screen. On the other, you can't possibly use raytracing as your only rendering technique for the amount of details in AAA video games. There's a reason we are not doing that now. New hardware more tailored for that kind of work will help, but don't expect the next GTA (or whatever) to have a full raytracing rendering like you see in those demo. Noticed the tiny size of the scenes? We'll see what the next generation of consoles brings us - especially since current gen are using AMD hardware and they've been really silent about this DXR announcement.

It's important because of internal websites stuck in the past. I doubt you'll ever need web assembly support for those.

The premise of matching modules and package one-to-one is I think incorrect. The way I see it, a package is just a library with a description of its dependencies, and possibly an already compiled binary. It could contains multiple modules. Plus, you probably need to link with a specific package version, and the module proposal doesn't have any way to specify a module version. So you can't just extract package dependencies from module usage.

Gamedev is a good candidate. Compiling in parallel, running different processes that transform resources, running a not yet optimized game engine in debug build, etc.

Yea. 60fps is tricky though, because the CPU is not twice as fast. With that GPU power you could render at 60fps at 1080p, but your game simulation might have a hard time reaching that speed.

You can get "true 4k" out of the console, but rendering at 4K is so taxing you'll chew all the extra processing power out of that GPU. That means that if developers want to increase rendering quality other than spatial resolution, they will probably resort to checkerboard rendering and/or dynamic resolution, which they needed to do on PS4 Pro. Of course, overall, you can probably expect better quality than a PS4 Pro, since there's more power. It's just not very wise to use that to output at 4k, especially when most people will have a hard time noticing the difference at living room distance from their TV.

Yea. It's not very hard - if you want something responsive, you have a limited amount of time between when you read input and when you present something on screen.

This delta can be greater than the rate at which you present images. For example, most game engines take 30ms to read input and simulate on a thread, then take 30ms to renders on another thread while the next frame is simulated. This means that they present something every 30ms, but the delay between input and rendering is 60ms. This delay can be up to 100ms and not be noticed by players.

Well, it's missing the interactivity part.

I think the pain point of DX12 is it's incompatibility with Windows 7 (and 8). However, Windows 10 is getting a lot of traction and has surpassed Windows 7 in gamers market share (see http://store.steampowered.com/hwsurvey/directx/).

With that in mind, if you are working on a AAA game that will come out in a couple of years, your target platforms will most likely be PS4, Xbox One and Windows 10. With those three, Vulkan doesn't make sense; DX12 will be used on Windows and Xbox One, and PS4 has its own proprietary API. Why complicate things with another API if you can use (almost) the same implementation for Xbox One and Windows 10?

If Windows 7 proves itself to be still somewhat popular amongs gamers, then Vulkan might be used to port games to this platform, since it's similar to DX12 so it will be way easier than a DX11 port.

However, right now, if game developers wants to use the latest API, Vulkan is a good choice since you will still be able to target Windows 7. If a lot of big games are released with Vulkan within 2 years, then it might slow down Windows 10's adoption and the market share of Vulkan systems (Windows Vista and up) will stay greater than the DX12 one (Windows 10) for longer. But for how many years? At one point, Windows 10 (or whatever's after) will dominate and the backward compatibility of Vulkan will stop being a useful marketing point.

So I'm not sure why you say there's no reason to use DX12. There's many. There's also a lot of good reasons to choose Vulkan that I didn't talk about here. I just don't see how Vulkan can dominates the Windows market, much like OpenGL.

This is badly phrased but it's for New Tweets only.

This is a great application of generally accepted software engineering principles (and some less common like "cleanup week"), but it would have been nice to see some real metric instead of some napkin graphs.

This is just an idea, since I never built such filter, but you could automate a large part of filtering NSFW images. A quick search on google lead to this paper : http://cs229.stanford.edu/proj2005/HabisKrsmanovic-ExplicitI... Once you have that in place, I guess it's better to make it agressive and report false positive as NSFW.

Google "safe image search" has the additional help of searching the content of the page the image is used. You might be able to do the same, up to some limit, by checking the http referer header field to know where requests are coming from. You could scan the referer's page for some keywords. This might give you a better idea of the context where the image is used. Note that this might be tricky, since you probably don't want traffic coming out of your server to some child porn site.

That said, those are just some ideas. Youtube has a good community that flags videos, but also an army of reviewer that look at the flagged content.

http://mobile.nytimes.com/2010/07/19/technology/19screen.htm...

Another way to look at it would be to try to manually select some images as "front page worthy", instead of trying to filter the bad stuff.

Exactly, and in fact it's usually not as simple as "you just work on the previous frame's data". There's a bunch of data dependencies that needs to be respected during a frame update. For example, if you update the player position wrt user input, you will want to update its animation in the same update, and also check if its not coliding with anything. Otherwise you're just lagging behind.

The advantage is that it's usable in a soft-real time scenario because it manage memory is a deterministic manner with ref counts instead of a GC.

I don't want an hoverboard. I want an hovercar. Replace roads with a sheet of copper and put grass over it.

When I frist saw the title of Zach Laine's talk, I was confused too. Type erasure for me is what happens with generics in Java, i.e. type information is discarded after compilation. In this case, the type is still present at runtime. I saw this technique described as "type safe void*", which is paradoxal to me. However, I do get the idea behind this, but I think "type erasure" is a wrong name for it.

Is it me or this feels terribly wrong?

My guess is it's to prevent "copycat" websites like those for stackoverflow.

I think the primary reason is just that when you play an FPS, you are looking at a screen, which is the same as looking at the TV when you watch a movie. In this real world setting, your brain is used to see all those effects.

A FPS for a VR headset however should try to mimic the eye instead, because I think it'll be more apparent that those effects shouldn't be there.