Author |
Message |
SoraK05
Joined: Fri 26 Oct 2012, 14:47:06 Posts: 81
|
Re: emulator efficiency
The more steps a program uses to access ram, the more the ram is used and the more it degrades as opposed to less steps for the identical results (where applicable). Accessing the hardware less by less steps and also less voltage will reduce overall degradation and consume less power. Thanks for the link to sns It would however be great to have an algorithm file based on a complete cycle accurate emulator to sift and optimize/shortcut Also, as for the computer doing math <arithmetic / logic>: http://www.computerhope.com/issues/ch000984.htmhttp://www.computerhope.com/jargon/e/edsac.htmMore direct math steps = less steps and less cycles/requirements.
|
Mon 02 Jun 2014, 22:57:06 |
|
|
Covarr
Screw y'all
Joined: Tue 28 Dec 2010, 08:27:37 Posts: 1147
|
Re: emulator efficiency
Dictionaries do not dictate reality. How the word may be defined, either currently or historically, has no bearing on what it actually does. Using definitions or history to discuss how computer science works is putting the cart before the horse.
_________________
|
Mon 02 Jun 2014, 23:07:42 |
|
|
jchadwick
Joined: Tue 21 Feb 2012, 05:42:15 Posts: 2564
|
Re: emulator efficiency
Still no progress being made. You have a fundamental misunderstanding or two about emulation that is going to require you to actually jump in and successfully write an emulator to solve.
_________________ "It's easy to win forgiveness for being wrong; being right is what gets you into real trouble." --Bjarne Stroustrup
|
Mon 02 Jun 2014, 23:30:36 |
|
|
Screwtape
Board Admin
Joined: Sat 11 Apr 2009, 04:21:58 Posts: 4783 Location: Australia
|
Re: emulator efficiency
Excuse me, sir, I find this image HIGHLY OFFENSIVE! I think I speak for everybody present when I say that we use SQUARE pixels around these parts - maybe 8:7 at the very most - but none of this 1:2 rubbish!
_________________ Maintainer of the unofficial git repository for bsnes.
The ending of the words is ALMSIVI.
|
Mon 02 Jun 2014, 23:39:43 |
|
|
neologix
Joined: Mon 06 Sep 2010, 07:50:15 Posts: 576
|
Re: emulator efficiency
This thread shall now be about pixel aspect ratio. Anyone else in the 16:10 camp?
|
Tue 03 Jun 2014, 00:29:11 |
|
|
jchadwick
Joined: Tue 21 Feb 2012, 05:42:15 Posts: 2564
|
Re: emulator efficiency
Did you mean: 8:5?
_________________ "It's easy to win forgiveness for being wrong; being right is what gets you into real trouble." --Bjarne Stroustrup
|
Tue 03 Jun 2014, 00:35:48 |
|
|
byuu
Joined: Fri 10 Apr 2009, 15:00:08 Posts: 13668
|
Re: emulator efficiency
1:0.625
Quit using marketing tactics with the big numbers.
|
Tue 03 Jun 2014, 01:46:22 |
|
|
CaptainJistuce
Joined: Wed 06 May 2009, 04:13:19 Posts: 4543
|
Re: emulator efficiency
This is completely wrong. Diodes, capacitors, resistors, et al do not wear with use. And the only true pixel aspect is 1/∞. Once it leaves RAM, it ceases being an ideal pixel and becomes corrupted by your "physics" and your "euclidean geometry"
_________________ This post best viewed at 800x600 ;write ! ! !
|
Tue 03 Jun 2014, 03:31:55 |
|
|
SoraK05
Joined: Fri 26 Oct 2012, 14:47:06 Posts: 81
|
Re: emulator efficiency
Hardware degrades - monitors don't have the same clarity / colour vibrancy after years, higher voltage and more use can make it more susceptible to damage, etc. http://ubuntuforums.org/showthread.php?t=488251CPU/Motherboards tend to last longer than the other components, however to be lighter on this hardware and the other components, as well as electricity/voltage optimizations can extend lifespan and usage of a program on a device powered by battery. In the long run it means longer time and usage before changing components, as well as run time with battery, and lower requirements. Normally electrical overheating is what causes more strain on hardware (like CPU/Motherboards), and while older hardware is generally unaffected newer hardware where overclocking is possible can lead to more heat/degradation and spikes can also affect hardware. With power cuts / spikes I have witnessed many pieces of hardware go, and keeping lower clocks can assist in avoiding this. Newer motherboards are generally more equip for such a thing.
|
Tue 03 Jun 2014, 03:49:26 |
|
|
The Doug
Joined: Wed 08 Sep 2010, 17:05:03 Posts: 133
|
Re: emulator efficiency
You are wrong. Sora only uses "non-euclidian geometry". To him there is no such thing like only one electron parallel to each other.
|
Tue 03 Jun 2014, 03:49:56 |
|
|
Braintrash
Joined: Wed 09 Feb 2011, 13:29:34 Posts: 425
|
Re: emulator efficiency
Hey, I am today throwing a 16/10 today in my JAMMA test station. ^^
|
Tue 03 Jun 2014, 06:58:22 |
|
|
SoraK05
Joined: Fri 26 Oct 2012, 14:47:06 Posts: 81
|
Re: emulator efficiency
Also, regarding an earlier post mentioning porting/rewriting a game/ROM to suit a specific platform, once the opcodes and shortcuts etc have been done in appropriate asm it is possible to patch a ROM directly to suit the opcodes and an overall emulator framework, and adjust any offset information met during instructions to have a ROM that will load as an EXE and have a requirement in a range close(r) to 20Mhz like the SNES <maintaining accuracy for output and using shortcuts which will break cycle accurate for measurement of exact expected cycle/hardware state data> An overall cycle accurate algorithm to transform to a shortcut version to use for asm and then this above would be ideal, and this can work on 486/PI and be very light with battery on something portable EDIT: On a relatively separate note, steps to optimize game code itself by adjusting of opcodes and data would be more of a final step to lower requirements more, which for some games could drop requirements to few Mhz overall
|
Tue 03 Jun 2014, 10:28:20 |
|
|
CaptainJistuce
Joined: Wed 06 May 2009, 04:13:19 Posts: 4543
|
Re: emulator efficiency
Monitors are very different than RAM and CPUs. While we're on it, LCD monitors and CRT monitors degrade significantly differently from each other, too. One can even be restored. But again, this is irrelevant as monitors are very different than CPUs and RAM and other ICs. Battery life is valid. But low-load does not extend the life of a CPU, limited use does not make RAM work for longer, and pushing fewer polys doesn't reduce the wear on a GPU. Overclocking was possible on older hardware too. And if you have overheating problems, you need to fix your cooling, not bitch because you overclocked your system and now it overheats if you load it too far(which defeats the point of overclocking if you suddenly have to limit your performance artificially because of it). Voltage spikes CAN damage hardware, but are unrelated to your processing load so much as they are your electrical supply.
_________________ This post best viewed at 800x600 ;write ! ! !
|
Tue 03 Jun 2014, 11:36:20 |
|
|
Kakashi
Joined: Mon 20 Apr 2009, 08:11:50 Posts: 5266 Location: 日本
|
Re: emulator efficiency
His explanation is to continue getting interesting. This is pretty sure it's a troll just to make the back and forth to free online translator. (His description keeps getting funnier. Pretty sure this is just a troll running things back and forth in an online translator.)
_________________ CaptainJistuce: He's totally in the wrong, Kakashi's 100% in the right. Note: The above statement is subject to act of byuu.
|
Tue 03 Jun 2014, 12:14:50 |
|
|
Kawa
Joined: Sun 05 Sep 2010, 15:42:48 Posts: 1344
|
Re: emulator efficiency
_________________ http://helmet.kafuka.org
|
Tue 03 Jun 2014, 12:29:55 |
|
|