Poster | Thread |
PR
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 17-Feb-2015 10:01:37
| | [ #21 ] |
|
|
 |
Super Member  |
Joined: 1-Sep-2004 Posts: 1962
From: Suomi-Finland | | |
|
| How can I measure i7 and OS4 G4? |
|
Status: Offline |
|
|
Samurai_Crow
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 17-Feb-2015 12:47:04
| | [ #22 ] |
|
|
 |
Elite Member  |
Joined: 18-Jan-2003 Posts: 2320
From: Minnesota, USA | | |
|
| @Asymetrix
Here's a YouTube vid of Doom running on ECS resolution: Click here!  |
|
Status: Offline |
|
|
pavlor
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 17-Feb-2015 15:41:43
| | [ #23 ] |
|
|
 |
Elite Member  |
Joined: 10-Jul-2005 Posts: 9673
From: Unknown | | |
|
| @PR
Quote:
How can I measure i7 and OS4 G4? |
Application benchmarks of the same (similar...) application versions. Eg. MPlayer, lame, dnetc, Doom, Quake etc.
From my experience Core i5-2500K 3.3 GHz is able to emulate 500-1000 MHz PowerPC CPU in WinUAE/QEMU (604e class - except FPU of course). Petunia JIT offers 1/5 of native performance - that is 200+ MHz 68060 under double emulation (QEMU-Petunia). So i7 is far more powerful than G4 (10x in single core performance).  |
|
Status: Offline |
|
|
klx300r
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 17-Feb-2015 15:54:55
| | [ #24 ] |
|
|
 |
Elite Member  |
Joined: 4-Mar-2008 Posts: 3857
From: Toronto, Canada | | |
|
| Quote:
apparently a compatible fpu unit is already available but does not fit into vampires fpga. no full scale mmu is planed afaik. |
ouch! fpu & mmu are a must for me _________________ ____________________________ c64-2sids, A1000, A1200T-060@50(finally working!),A4000-CSMKIII ! My Master Miggies- Amiga 1000 & AmigaOne X1000 ! mancave-ramblings X1000 I BELIEVE 
|
|
Status: Offline |
|
|
PR
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 17-Feb-2015 18:39:13
| | [ #25 ] |
|
|
 |
Super Member  |
Joined: 1-Sep-2004 Posts: 1962
From: Suomi-Finland | | |
|
| I know i7 is quite powerfull vs any old G4 etc. In a Happy land even would not care. Offcourse there is Altivec and in the Motorola -060 a Math prosessor.
Just wanted a program for curiosity to test the Mips for comparison.
Got a Military Class Motherboard for the fastest things ever. i7 to be upgraded, DD3 too, Graphics card is not yet even made so have to settle with the NVidia.
It's like a rocket. This vs the XE...
- First it was Kb, then MB now GB. TB or two is ok for now;)
(No Filebytes or casette) Last edited by PR on 17-Feb-2015 at 06:50 PM. Last edited by PR on 17-Feb-2015 at 06:48 PM. Last edited by PR on 17-Feb-2015 at 06:43 PM.
|
|
Status: Offline |
|
|
kolla
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 18-Feb-2015 4:55:37
| | [ #26 ] |
|
|
 |
Elite Member  |
Joined: 20-Aug-2003 Posts: 3376
From: Trondheim, Norway | | |
|
| Yeah, I certainly want a compatible MMU too. Or to be precise - I would very much pefer a drop in replacement for an existing CPU, such as a MC68060, or else a lot of software simply wont work, including a few operating systems. Also, if this is "full 020 compatible", then why does it say 68EC020? _________________ B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC
|
|
Status: Offline |
|
|
wawa
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 18-Feb-2015 11:01:18
| | [ #27 ] |
|
|
 |
Elite Member  |
Joined: 21-Jan-2008 Posts: 6259
From: Unknown | | |
|
| any other particular use for mmu than muforce? mac emulation? |
|
Status: Offline |
|
|
Hypex
 |  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 18-Feb-2015 13:13:19
| | [ #28 ] |
|
|
 |
Elite Member  |
Joined: 6-May-2007 Posts: 11351
From: Greensborough, Australia | | |
|
| @pavlor
Is that 68060 running as a JIT under another JIT? I thought that would surely break as a JIT² would multiply the chances of crashing
I ran PC Task Turbo under Petunia once and it just crashed it. Last edited by Hypex on 19-Feb-2015 at 12:54 PM.
|
|
Status: Offline |
|
|
OlafS25
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 18-Feb-2015 16:01:46
| | [ #29 ] |
|
|
 |
Elite Member  |
Joined: 12-May-2010 Posts: 6480
From: Unknown | | |
|
| a lot of software needs MMU?
What for example? |
|
Status: Offline |
|
|
pavlor
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 18-Feb-2015 19:23:42
| | [ #30 ] |
|
|
 |
Elite Member  |
Joined: 10-Jul-2005 Posts: 9673
From: Unknown | | |
|
| @Hypex
Quote:
Is that 68060 running as a JIT under another JIT? |
68060 was only for performance comparison. Yes, QEMU JIT emulates PowerPC CPU and Petunia JIT in OS4 emulates 68k CPU - it is stable in my experience (ADoom 68k and other 68k demanding games/applications work OK). |
|
Status: Offline |
|
|
pavlor
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 18-Feb-2015 19:26:09
| | [ #31 ] |
|
|
 |
Elite Member  |
Joined: 10-Jul-2005 Posts: 9673
From: Unknown | | |
|
| @wawa
Quote:
Used for screen refresh, not required (ShapeShifter works great with WinUAE/OS3/JIT configuration).
MMU is certainly usebale for developers. From 68k user perspective, it doesn“t bring many benefits. |
|
Status: Offline |
|
|
Niolator
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 18-Feb-2015 22:57:40
| | [ #32 ] |
|
|
 |
Super Member  |
Joined: 3-May-2003 Posts: 1420
From: Unknown | | |
|
| @Rob
Quote:
Quote:
Quote: 62MIPS... Doesn't a 060/50 do about 60MIPS?
|
I think it's closer 50MIPS. What's really remarkable here is that the Vampire 600 card this is running on is significantly cheaper than any 68060 board you can get your hands on. |
In that case a purchase might be warranted for me. |
|
Status: Offline |
|
|
kolla
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 19-Feb-2015 2:15:11
| | [ #33 ] |
|
|
 |
Elite Member  |
Joined: 20-Aug-2003 Posts: 3376
From: Trondheim, Norway | | |
|
| For example, any kind of virtual memory, like VMM, will not work without MMU, which some of us would very much need since the ammount of RAM is so limited. Other software relevant for me is Linux/m68k, I would love new fast m68k hardware to run on, it would hugely benefitial for embedded communitues that a FPGA m68k is very compatible with existing hardware, in all aspects. Let me just say that a company, Cose Sourcerers in Australia, were paid to bring Linux, glibc and the gcc toolchain up to date on m68k a few years ago - not just for ColdFire, but also for m68k in general. There are a few commercial FPGA m68k softcores available in the embedded industry, plenty of legacy systems that still rely on m68k.
Anyways, I expect that more capable alternatives will show up, just like WinUAE didn't have MMU before it was back ported from the atari emulator Aranym. _________________ B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC
|
|
Status: Offline |
|
|
NovaCoder
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 19-Feb-2015 5:41:24
| | [ #34 ] |
|
|
 |
Regular Member  |
Joined: 16-Apr-2008 Posts: 492
From: Melbourne (Australia) | | |
|
| Quote:
I think it's closer 50MIPS. What's really remarkable here is that the Vampire 600 card this is running on is significantly cheaper than any 68060 board you can get your hands on. |
Yep this is another very exciting development in the FPGA space.
People wouldn't have thought it was possible to make such a powerful A600 card at this kind of cost a few years ago.
Well done to all involved :)
Now, hurry up and use a bigger FPGA in a Vampire board for A1200/A3000/A4000  |
|
Status: Offline |
|
|
Deniil715
 |  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 19-Feb-2015 10:06:37
| | [ #35 ] |
|
|
 |
Elite Member  |
Joined: 14-May-2003 Posts: 4238
From: Sweden | | |
|
| @BigGun
Quote:
Lets use Phoenix as comparison to G4.
The 68K can do a lot more per instruction. |
The whole point of RISC has been diminished the last 10-15 years. The whole point was to make an architecture that did not need so many transistors by having fewer instructions (Reduced Instruction Set), and where every instruction was simpler requiring a shallower logic net, and again, fewer transistors. This would allow it to be clocked faster than CISC CPUs.
However, since technology has advanced into smaller and less energy-consuming transistors, there is no need to spare on the transistor count anymore, so instead of making small and shallow logic nets with simpler instructions to be able to clock high, we simply make a wider, but shallow, logic net with many transistors, and so we can run CISC code at the same high clock speed.
So now PPC, being a (rather complex) RISC CPU, needing to do many relatively simple instructions to get the job done, we have CISC (x86) running at the same speed by simply using a lot more transistors, running more complex instructions in its wide, but equally shallow logic net. PPC would have to clock over 10GHz to compete with x86, but that's not physically possible. The RISC idea became pointless. _________________ - Don't get fooled by my avatar, I'm not like that (anymore, mostly... maybe only sometimes)  > Amiga Classic and OS4 developer for OnyxSoft.
|
|
Status: Offline |
|
|
BigGun
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 19-Feb-2015 10:38:09
| | [ #36 ] |
|
|
 |
Regular Member  |
Joined: 8-Aug-2005 Posts: 438
From: Germany (Black Forest) | | |
|
| Quote:
The RISC idea became pointless. |
Developing a CISC CPUs give some challenges 1) Its difficult to decode long and complex CISC instructions quickly. 2) That all instructions can have different length makes decoding several instruction per cycle more difficult. 3) The testing effort for CISC CPU is very high
RISC offer simple development of the CPU 1) As the core is simpler, the testing can be done with less man power/time 2) As all instructions have the same length decoding multiply instructions is easy. All you need to do is copy 2 paste.
A CISC CPU can do more work with less instructions.
Decoding several instructions in a RISC CPU is relative simple. The main bottleneck for a RISC CPU is not the decoding but the dependency checking of several instructions per cycle. A RISC CPU needs to execute more instructions per cycle to do the same work. The effort of dependency checking is the same for CISC and RISC and here RISC is bottlenecked. _________________ APOLLO the new 68K : www.apollo-core.com
|
|
Status: Offline |
|
|
ChaosLord
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 19-Feb-2015 21:46:49
| | [ #37 ] |
|
|
 |
Cult Member  |
Joined: 4-Apr-2005 Posts: 782
From: Houston, Texas USA | | |
|
| Quote:
any other particular use for mmu than muforce? mac emulation? |
CodeProbe Source Level Debugger (from SASC) Bdebug (from BarFly) Numerous other debuggers CyberGuard ShapeShifter Emulation VMM GigaMem All other Virtual Memory solutions MuGuardianAngel MuProtectModules MuMove4k MuFastZero mmu.library MuFastRom MuFastChip MuEVD MuLink (This is the one that lets you write-protect your code hunks) MuLockLib MuMapROM MuRedox (!!!!) MuScan MuSetCacheMode MuOverlayManager MuOmniSCSIPatch Elbox Mediator PCI slots
And tons of other stuff I forgot about.
680x0 MMU is awesome! _________________ Wanna try a wonderfull magical Amiga strategy game? Total Chaos AGA
|
|
Status: Offline |
|
|
ChaosLord
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 19-Feb-2015 21:59:05
| | [ #38 ] |
|
|
 |
Cult Member  |
Joined: 4-Apr-2005 Posts: 782
From: Houston, Texas USA | | |
|
| Quote:
MMU is certainly usebale for developers. From 68k user perspective, it doesn“t bring many benefits. |
If you are a user who wants to report bugs to developers then an MMU is a wonderful and magically fantastic device to have.
If you are a user who simply enjoys Amiga software that actually works then you want every Amiga to have a cheap and fully functional MMU.
If you are a user who enjoys a stable and fast Amiga then you want an MMU. _________________ Wanna try a wonderfull magical Amiga strategy game? Total Chaos AGA
|
|
Status: Offline |
|
|
matthey
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 19-Feb-2015 23:23:57
| | [ #39 ] |
|
|
 |
Elite Member  |
Joined: 14-Mar-2007 Posts: 2522
From: Kansas | | |
|
| Quote:
And tons of other stuff I forgot about. |
With your long Mu collection list you somehow forgot MuForce. Enforcer is kind of old but probably deserves mention. The Fusion 68k MacOS emulator could also be up there.
Quote:
If you are a user who enjoys a stable and fast Amiga then you want an MMU. |
Protecting zero page certainly helps stability. I'm not sure how an MMU makes the Amiga fast though. |
|
Status: Offline |
|
|
ChaosLord
|  |
Re: Phoenix version of 200MHz 68020 available for testers Posted on 20-Feb-2015 1:39:01
| | [ #40 ] |
|
|
 |
Cult Member  |
Joined: 4-Apr-2005 Posts: 782
From: Houston, Texas USA | | |
|
| Quote:
I'm not sure how an MMU makes the Amiga fast though. |
You must be getting even older than me.
1. Copying the Kickstart ROM into FASTram.
2. MuRedox is a MuLib based "on the fly" speedup patch for 68040 and 68060 based Amiga boards. The 68040 and 68060 do not implement all instructions of the MC 68K family. The unimplemented instructions - mainly FPU instructions - generate an exception and need to be emulated by the 68040 resp. 68060.library. This is the job of the so- called "FPSP routines" (floating point support package) within the CPU libraries. MuRedox detects these instructions as soon as they generate the emulator exceptions, runs a "just-in-time" compiler that generates a "stub replacement routine" for this specific instruction and patches the replacement routine into the running program. Hence, MuRedox replaces the overhead of the emulator trap on the next use of the same instruction sequence.
3. An MMU makes Mac Emulation massively faster. _________________ Wanna try a wonderfull magical Amiga strategy game? Total Chaos AGA
|
|
Status: Offline |
|
|