Topics

1

Quake / Speedruns & Defrags

« on: June 26, 2020, 04:36:52 PM »

Damned impressive 100% speedrun of Quake1 (must kill all enemies and find all secrets.)


https://www.youtube.com/watch?v=en1Ke7ZO_xk

2

Quake / Q2VKPT - Quake 2 real-time raytracing using RTX

« on: January 18, 2019, 09:55:27 AM »

As a research project, someone's added real-time raytracing to Quake2.

http://brechpunkt.de/q2vkpt/

Q2VKPT is the first playable game that is entirely raytraced and efficiently simulates fully dynamic lighting in real-time, with the same modern techniques as used in the movie industry (see Disney's practical guide to path tracing). The recent release of GPUs with raytracing capabilities has opened up entirely new possibilities for the future of game graphics, yet making good use of raytracing is non-trivial. While some games have started to explore improvements in shadow and reflection rendering, Q2VKPT is the first project to implement an efficient unified solution for all types of light transport: direct, scattered, and reflected light (see media). This kind of unification has led to a dramatic increase in both flexibility and productivity in the movie industry. The chance to have the same development in games promises a similar increase in visual fidelity and realism for game graphics in the coming years.
This project is meant to serve as a proof-of-concept for computer graphics research and the game industry alike, and to give enthusiasts a glimpse into the potential future of game graphics. Besides the use of hardware-accelerated raytracing, Q2VKPT mainly gains its efficiency from an adaptive image filtering technique that intelligently tracks changes in the scene illumination to re-use as much information as possible from previous computations. ↑

The project is released as open source on GitHub, integrating our Vulkan path tracer into the Q2PRO client. The project was spawned by the need for fast-paced test content in computer graphics research, and was motivated by the first intriguing results of an experimental path tracing renderer in 2016.

Why Quake II?

Since Quake II is open source and has a long-standing modding tradition, it is a great sandbox for putting academic research to the test in the real world. Particularly, the game has fast-paced action and is played competitively, setting high standards for the performance and robustness of any implemented rendering techniques.

Why are there no particles? I want my flies and rail gun sparks back!

Unfortunately, we ran out of time to add lights for the rail gun. As to other particles, the decision to fully raytrace everything in the game made rendering particles a bit less trivial. If we find the time, we might fix it in the future!

Note that ray-tracing is not fully realized yet across the board (e.g. reflective glass pane, muzzle flashes, etc.) but, the parts that work are at least intriguing:


https://www.youtube.com/watch?v=vrq1T93uLag

3

Quake / Quake done 100% Quickest

« on: June 03, 2017, 06:50:32 PM »

"'Quake done 100% Quickest' is a run through all of Quake on Nightmare skill, going through the levels as fast as possible, killing every single monster and finding every single secret in every single map of Quake."


https://www.youtube.com/watch?v=ZhzXKMqZBBc
<span data-s9e-mediaembed="youtube" style="display:inline-block;width:100%;max-width:640px"><span style="display:block;overflow:hidden;position:relative;padding-bottom:56.25%"><iframe allowfullscreen="" scrolling="no" style="background:url(https://i.ytimg.com/vi/ZhzXKMqZBBc/hqdefault.jpg) 50% 50% / cover;border:0;height:100%;left:0;position:absolute;width:100%" src="https://www.youtube.com/embed/ZhzXKMqZBBc"></iframe></span></span><br /><a href="https://www.youtube.com/v/ZhzXKMqZBBc" target="_blank" class="new_win">https://www.youtube.com/v/ZhzXKMqZBBc</a>

Only takes them 9 minutes to get through the original shareware campaign maps.

4

Science / "What is Technology Doing to Us?"

« on: April 30, 2017, 12:18:20 AM »

"What is Technology Doing to Us?"

https://www.youtube.com/watch?v=jlPF9_1VIso
<span data-s9e-mediaembed="youtube" style="display:inline-block;width:100%;max-width:640px"><span style="display:block;overflow:hidden;position:relative;padding-bottom:56.25%"><iframe allowfullscreen="" scrolling="no" style="background:url(https://i.ytimg.com/vi/jlPF9_1VIso/hqdefault.jpg) 50% 50% / cover;border:0;height:100%;left:0;position:absolute;width:100%" src="https://www.youtube.com/embed/jlPF9_1VIso"></iframe></span></span><br /><a href="https://www.youtube.com/v/jlPF9_1VIso" target="_blank" class="new_win">https://www.youtube.com/v/jlPF9_1VIso</a>

[talks about an insider perspective on competing armies of behavioral sociologists in a corporate arms race (to persuade us for every ounce of our available attention)]

5

Science / Lawrence Krauss - Cosmic Natural Selection: Darwin Day Lecture 2017

« on: April 27, 2017, 01:02:02 AM »

https://www.youtube.com/watch?v=v6ef99T-IJE
<span data-s9e-mediaembed="youtube" style="display:inline-block;width:100%;max-width:640px"><span style="display:block;overflow:hidden;position:relative;padding-bottom:56.25%"><iframe allowfullscreen="" scrolling="no" style="background:url(https://i.ytimg.com/vi/v6ef99T-IJE/hqdefault.jpg) 50% 50% / cover;border:0;height:100%;left:0;position:absolute;width:100%" src="https://www.youtube.com/embed/v6ef99T-IJE"></iframe></span></span><br /><a href="https://www.youtube.com/v/v6ef99T-IJE" target="_blank" class="new_win">https://www.youtube.com/v/v6ef99T-IJE</a>

This lecture really took off for me about 30 minutes in, around the point Prof. Krauss was explaining how Maxwell had calculated the rate of electromagnetic propagation, based on the both the measurement of the strength of two electric charges and the measurement of the strength of the force between two magnets; the speed of this propagation calculated by Maxwell being the speed of light!

A lot of good stuff. (Some audio issues early on eventually get ironed out.)

6

tastyspleen.net / east.tastyspleen.net (tourney4, et al.) was unexpectedly vaporized

« on: May 01, 2016, 02:52:34 PM »

Greetings,

So, the hosting company where t4 was located has deleted the entire datacenter where the server was hosted.

Response to my inquiry: "That device was hosted in our LDJ data center. That datancenter[sic] got closed last week. Hence, all devices associated to it were deleted."

As far as I can tell, this was done with no advance warning. (They claim they sent an email beforehand, but I see no record of it anywhere, including spam folders. I've got plenty of emails from them around that time saying things like, "The network team has informed us that network connectivity has been restored." But nothing whatsoever about them pulling the plug.)

Anyway... At this time, I have no immediate plans to seek other hosting. (Frankly I'm looking for ways to save money, so while this was unexpected, the $48/mo savings is not entirely unwelcome.)

7

0x1337c0de / 64K ought to be enough for anyone

« on: April 29, 2016, 08:43:40 PM »

Wow...

These are real-time video captures from a 64K executable: All graphics, sound, music are contained in the 64K exe.

These were part of a 64K demo category from a contest from earlier this year (March, 2016.)

I guess this has caught me slightly by surprise: I'm used to "Moore's Law" and the corresponding process shrinks yielding more transistors, more storage, etc.

But it hadn't occurred to me to think about how much more one could now get done in a 64K binary.


The following won 2nd place in the contest:

<span data-s9e-mediaembed="youtube" style="display:inline-block;width:100%;max-width:640px"><span style="display:block;overflow:hidden;position:relative;padding-bottom:56.25%"><iframe allowfullscreen="" scrolling="no" style="background:url(https://i.ytimg.com/vi/_SEGm2lUSY0/hqdefault.jpg) 50% 50% / cover;border:0;height:100%;left:0;position:absolute;width:100%" src="https://www.youtube.com/embed/_SEGm2lUSY0"></iframe></span></span><br /><a href="https://www.youtube.com/v/_SEGm2lUSY0" target="_blank" class="new_win">https://www.youtube.com/v/_SEGm2lUSY0</a>


This was the 1st place entry:

<span data-s9e-mediaembed="youtube" style="display:inline-block;width:100%;max-width:640px"><span style="display:block;overflow:hidden;position:relative;padding-bottom:56.25%"><iframe allowfullscreen="" scrolling="no" style="background:url(https://i.ytimg.com/vi/JZ6ZzJeWgpY/hqdefault.jpg) 50% 50% / cover;border:0;height:100%;left:0;position:absolute;width:100%" src="https://www.youtube.com/embed/JZ6ZzJeWgpY"></iframe></span></span><br /><a href="https://www.youtube.com/v/JZ6ZzJeWgpY" target="_blank" class="new_win">https://www.youtube.com/v/JZ6ZzJeWgpY</a>


Heavy use of procedural pixel shader techniques, etc., etc., but goddamn.

Packing that all into a 64K binary... 

8

vanilla / avoiding giant maps (unless lots of players)

« on: October 05, 2015, 02:33:08 AM »

Salutations,

Finally coded wallfly to observe the number of active players, , and skip maps coded to be too large for the active player count.

 

(still working on the same for mutant... got thru 200+ mutant maps today, something like 550+ still to go  )

But anyway, if there are only a few 'active' players on the server, smaller maps will be used. (Similar logic now also affects the behavior of commands like `mymap random`. It will try to pick from maps suitable to the player count.)

'Active' players are determined by players who got a frag recently. (So if a server is full if spectators / idle players, they don't weigh into the decision. Just the players actually fragging.)

Anyway, might require some tweaking/tuning over time, but hopefully the server should be less likely to empty out due to scheduling maps whose vastness exceeds the player count.

9

tastyspleen.net / WallFly server maintenance reboot @ 3 AM CDT, Sunday May 24

« on: May 23, 2015, 11:54:26 AM »

Salutations!

The server that runs WallFly has been up for 1604 days since its last reboot, but it's developing errors with some of its RAM modules.

nexus kernel: [137102675.670956] EDAC MC0: UE row 3, channel-a= 2 channel-b= 3 labels "-": (Branch=1 DRAM-Bank=2 RDWR=Write RAS=12103 CAS=0 FATAL Err=0x4 (>Tmid Thermal event with intelligent throttling disabled))

nexus kernel: [138131887.882526] EDAC MC0: UE row 1, channel-a= 2 channel-b= 3 labels "-": (Branch=1 DRAM-Bank=3 RDWR=Write RAS=10635 CAS=0 FATAL Err=0x1 (Alert on non-redundant retry or fast reset timeout))

etc.

The server will be shut down for maintenance at about 3 AM CDT tomorrow morning (Sunday, May 24.)

Hopefully back up and running within an hour or so.

10

Quake / Per-map FPH stats

« on: March 08, 2015, 06:48:37 PM »

Finally wrote a parser to pull FPH stats from FFA server logs:

http://tastyspleen.net/quake/stats/frags-fph/


Currently these go back to 2008 (quirks in the logfile format prior to then require alternate parsing.)

11

Science / The AI Revolution: Superintelligence, Immortality, Extinction

« on: February 05, 2015, 02:20:36 AM »

Part 1: The AI Revolution: The Road to Superintelligence

Part 2: The AI Revolution: Our Immortality or Extinction









For me, an interesting article overall.

(I suspect some physical limits relating to information processing, and impediments like the General Halting Problem (and perhaps the existence of NP-complete problems), may place some constraints on the rate of takeoff from AGI to ASI... And I also suspect the 'quality' of intelligence may not be able to be ratcheted up to the tremendous multipliers postulated in the article; -- I don't know what sort of law might govern that, but I'd be surprised if there didn't work out to be some sort of diminishing returns, where even a superintelligence accustomed to designing the next smarter version of itself might begin finding it increasingly difficult to make incremental gains. Nevertheless, I've little doubt there's room to be many times smarter than humans before hitting those sorts of limits, if they exist.)

12

tastyspleen.net / Unplanned Server Outage - 21aug2014

« on: August 21, 2014, 10:59:38 PM »

Salutations,

Here's a synopsis of the unplanned server outage that occurred today.

At 1330 days of uptime, our primary server fragbait.tastyspleen.net crashed with a kernel panic in the Ethernet driver.

Subsequent attempts to reboot and/or power cycle the server were met with failures to bring up the network driver:

[    1.921773] e1000e: Intel(R) PRO/1000 Network Driver - 1.0.2-k2
[    1.921836] e1000e: Copyright (c) 1999-2008 Intel Corporation.
[    1.921945] e1000e 0000:04:00.0: PCI INT A -> GSI 18 (level, low) -> IRQ 18
[    1.922025] e1000e 0000:04:00.0: setting latency timer to 64
[    1.922140] e1000e 0000:04:00.0: irq 54 for MSI/MSI-X
[    2.172397] e1000e 0000:04:00.0: PCI INT A disabled
[    2.172467] e1000e: probe of 0000:04:00.0 failed with error -2
[    2.172552] e1000e 0000:04:00.1: PCI INT B -> GSI 19 (level, low) -> IRQ 19
[    2.172629] e1000e 0000:04:00.1: setting latency timer to 64
[    2.172754] e1000e 0000:04:00.1: irq 54 for MSI/MSI-X
[    2.423008] e1000e 0000:04:00.1: PCI INT B disabled
[    2.423070] e1000e: probe of 0000:04:00.1 failed with error -2


After I opened a support ticket, I was able to watch the tech assigned by SoftLayer over the virtual KVM as he attempted to troubleshoot.

Sometimes when the tech would boot into a "rescue" kernel, the network driver would load. Sometimes after that when he booted back into the fragbait kernel the networking would still be OK--which is why we had a brief interval in the afternoon where the system was back up for awhile. However, on subsequent reboots the networking would go back to the error condition above.

The tech suggested updating the BIOS on the motherboard, but unfortunately this had no noticable effect on the problem.

The tech then continued running various diagnostic programs at occasional intervals, however the programs the tech was running depended on the Ethernet devices existing (but the devices did not exist, because the driver was unable to load, again related to the error above -- so these diagnostic tools were unable to run.)

So it was an interesting experience watching sometimes lengthy periods of inactivity punctuated by the running of another diagnostic tool which had no chance of working.

Eventually the problem was escalated to a higher tier tech who was aware of a rather bizarre problem with the particular motherboard on the fragbait server.

He was able to get the networking back up, and explained the issue as follows:

I am going to go ahead and put an account note on your account so that we can get this fixed in a faster fashion in the future.

There is a specific issue with the motherboard in this server. There is a trick of sorts that has to be performed as this motherboard can lose network access at times.

basically for the server fragbait2 which has the X7DBR-E we have to shutdown the machine, set the network interfaces to 0 then push the config to our switches.

Then wait a minute or two, then power back on the machine and then reset the speed of the interfaces back to the required speed which for this one is 1Gbps.

After that networking starts to work and is a documented issue for this board.

I did go ahead and add the notes for this server so hopefully any issues will be fixed in a quick manner.

So.... that's definitely an odd one. I'm glad somebody there knew a workaround. It will be interesting to see how long it lasts. (If it can last another 1330 days, I won't complain much. But who knows...)

13

/dev/random / DOOM 20th anniversary playthrough with John Romero

« on: July 13, 2014, 01:41:14 AM »

Chill 2013 interview with John Romero playing co-op with the interviewer through the DOOM shareware maps, with plenty of commentary about the level design:

<span data-s9e-mediaembed="youtube" style="display:inline-block;width:100%;max-width:640px"><span style="display:block;overflow:hidden;position:relative;padding-bottom:56.25%"><iframe allowfullscreen="" scrolling="no" style="background:url(https://i.ytimg.com/vi/b86PwuzyxlY/hqdefault.jpg) 50% 50% / cover;border:0;height:100%;left:0;position:absolute;width:100%" src="https://www.youtube.com/embed/b86PwuzyxlY"></iframe></span></span><br /><a href="https://www.youtube.com/v/b86PwuzyxlY" target="_blank" class="new_win">https://www.youtube.com/v/b86PwuzyxlY</a>

Example: 39:15 -- "...understand the 3D design of the level -- by making them come through it multiple times, is from Zelda, actually. Because, at the beginning of Zelda 1 on the NES, the first area that you start in, has these bushes that--later--you get a lantern that you can throw fire with: And when you burn down those bushes, there's a dungeon under it! And that dungeon was in the first room of the game! And to me I was like, wow, that's really cool how they re-used the original area. So I always try and do that when I'm designing: how do I maximize my design by re-using it multiple times in different ways?"

Lots more commentary like the above...




P.S. 57:40 -- definitive 'gib' pronunciation 

14

Science / Cosmos: A Spacetime Odyssey

« on: March 16, 2014, 04:24:04 AM »

First episode of the new Cosmos series is online here:

http://www.cosmosontv.com/watch/183733315515  (Says: Expires in 49 days :S )

Really nicely done! I hope it gets shown in a lot of classrooms. The "cosmic calendar" metaphor was a clever way to bridge cosmic time-scales to a familiar reference point.

Looking forward to episode 2...

15

Jokes / Pseudoscientific Audiophile Reviews

« on: March 10, 2014, 10:52:17 AM »

"Pseudoscientific wank from the pages of audiophile magazines."

http://wathifi.tumblr.com/