Questions About rpcs3
Started by rafaPS360gamer




15 posts in this topic
ssshadow
Administrator
*******


4
2,494 posts 63 threads Joined: Aug 2017
02-03-2014, 09:12 PM -
#9
(02-03-2014, 08:58 PM)hlide Wrote:
(02-03-2014, 07:31 PM)ssshadow Wrote: Yes, and you can just use Moores law to estimate when phones/tablets would be fast enough.
oh god! that Moore law has not been valid any longer for a decade now. Look at the actual frequencies of intel and amd cpus, they are stuck now. It is not a coincidence that they try the multi-core roadmap instead. Unless they could find a great new kind of battery to feed a hungry handheld and keep it cool as much as possible without any fan or watercooling, you cannot expect the same thing as for a pc. My opinion is ps3 emulation will be history by the time handhelds will be able to compete with the current best pc.

Moore's law as commonly refereed to has nothing to do with clock frequency, but rather the number of transistors on a chip. Even if IPC does not improve, smaller transistors will give you more cores on the same die area, which does give you more performance at the same voltage (for certain applications, probably not this one, but whatever, this is generally speaking.).

And also, Moore's law has not been invalid for the last decade, it was in fact completely valid until at least 2011 and Intel already have 14 nm fabs ready for the next generation, so it all looks good to me. And BTW, even current Haswell Core processors at 22 nm can bee cooled passively in Windows 8 tablets, with great performance. Broadwell will probably be even better in this regard, while improving performance further.

Let us look at the passive Haswell i5 further. It will run at a minimum of 1,6 ghz, is a dual core, and has 3 MB of cache. A desktop i5 is at about 3 ghz (base speed), is a quad core, and has 6 MB of cache. Both are of the same architecture. Let's say half the number of cores is half the performance, and half the clock speed is half the performance again. This is actually not that bad. If a desktop could emulate something at 60 fps, the tablet would be at 15. Unplayable, sure, but give it a simpler game, and a couple of years of performance improvements, and the tablet would not be that bad.

And this is assuming half the cores is half the performance, with for instance pcsx2 this is usually not the case.


Messages In This Thread
RE: Questions About rpcs3 - by rafaPS360gamer - 02-01-2014, 05:05 PM
RE: Questions About rpcs3 - by Zekro - 02-01-2014, 05:34 PM
RE: Questions About rpcs3 - by rafaPS360gamer - 02-01-2014, 11:28 PM
RE: Questions About rpcs3 - by ssshadow - 02-02-2014, 04:57 PM
RE: Questions About rpcs3 - by hlide - 02-03-2014, 01:22 PM
RE: Questions About rpcs3 - by rafaPS360gamer - 02-03-2014, 06:53 PM
RE: Questions About rpcs3 - by ssshadow - 02-03-2014, 07:31 PM
RE: Questions About rpcs3 - by hlide - 02-03-2014, 08:58 PM
RE: Questions About rpcs3 - by ssshadow - 02-03-2014, 09:12 PM
RE: Questions About rpcs3 - by hlide - 02-03-2014, 09:41 PM
RE: Questions About rpcs3 - by ssshadow - 02-03-2014, 09:52 PM
RE: Questions About rpcs3 - by hlide - 02-03-2014, 10:11 PM
RE: Questions About rpcs3 - by rafaPS360gamer - 02-03-2014, 11:14 PM
RE: Questions About rpcs3 - by Tallu - 02-13-2014, 12:49 AM
RE: Questions About rpcs3 - by Aishou - 02-13-2014, 02:20 AM
RE: Questions About rpcs3 - by rafaPS360gamer - 02-24-2014, 07:14 PM

Forum Jump:


Users browsing this thread: 1 Guest(s)