From AMCC web site:

https://www.appliedmicro.com/MyAMCC/jsp/public/productDetail/product_detail.jsp?productID=PPC440EP

PowerPC 440EP - CPU Performance 333-667MHz - Features DDR1 Memory Controller, FPU, GPIO, PCI, USB 1.1, USB2.0

Product Status: Production Release


Advanced technology for Imaging and Industrial Control applications

General Description
The PowerPC 440EP embedded processor offers exceptional performance, design flexibility, and robust features geared to demanding Imaging, industrial control, networking and other embedded applications. With speeds of up to 533MHz, PowerPC Book E Architecture, and a rich peripheral mix, PowerPC 440EP processors are ideally suited for a wide range of high-performance applications.

AMCC's advanced 0.13-micron CMOS copper process technology and an innovative design provide low power dissipation and a small footprint. Application code compatibility with other PowerPC processors enables manufacturers to bring products to market quickly and at a low cost to satisfy changing needs. High integration with robust peripheral support can further simplify board design and help reduce manufacturing costs.

Highlights

Delivers 333MHz to 533MHz performance for high-speed embedded designs 32-bit implementation of the Book E Enhanced PowerPC� Architecture Superscalar PowerPC 440 core, with large L1 caches, with the high-speed IBM CoreConnect� bus technology Single/double-precision floating point unit State-of-the-art peripherals including DDR SDRAM, and extensive connectivity via Ethernet, USB, UARTs, IIC, SPI and PCI Offers low power dissipation and small form factor for high-density and power-conscious applications Provides application code compatibility with other PowerPC processors

Features
440 core and DDR interface connect to CoreConnect PLB4
100-133MHz, 128-bit data read and write buses, 36-bit address bus Performance monitor Bridge to PLB3, 100-133MHz, for PCI, EBC, Ethernet, and OPB peripherals

On-chip Double Data Rate (DDR) SDRAM controller
32-bit interface with optional ECC 13-bit addressing 1.1 GBps peak data rate Support for 4 banks of up to 256MB, maximum capacity of 1GB Support for 64, 128, 256, and 512Mb DDR devices, with CAS latencies of 2, 2.5, or 3

PCI interface
32-bit PCI V2.2, with 3.3V interface, at frequencies of up to 66MHz Multiple read prefetch and write post buffers Ability to boot processor from PCI bus memory

USB
USB 1.1 host, MAC and PHY USB 2.0 Device MAC USB 2.0 device MAC UTMI or USB 1.1 device PHY�device supports 6 end points (3 in, 3 out), 1024 Byte FIFO (double buffering of 512 byte packets)

FPU
5 stage FPU with 2.0 MFLOPS/MHz (SP/DP) Hardware support for IEEE 754 Single-precision and double- precision operation Single cycle throughput on most instructions Thirty-two 64-bit floating point registers

External bus controller
50-66MHz 8-bit or 16-bit external data bus width Up to 30-bit address bus Support for up to 6 ROM, EPROM, SRAM, Flash, or slave peripheral I/O devices External master support

Nand Flash controller
Block-oriented device, accessed in a manner much like diskette drives with pages, blocks, and, in some devices, zones or districts 1 to 4 banks of Nand Flash supported on EBC Direct interfacing to: Discrete Nand Flash devices (up to 4 devices) and SmartMedia Card socket (22-pins) Device sizes 4MB-256MB supported 512-byte + 16-byte or 2KB + 64-byte device page sizes supported Boot-from-Nand: Execute a linear sequence of boot code out of single page of 1st block (512- bytes) Support DMA to allow direct, no-processor-intervention block copy from Nand Flash out to SDRAM - ECC providing single bit error correction and double bit error detection in each 256-bytes of stored data Chip selects shared with EBC

DMA controller
4 independent channels supporting internal and external peripherals

---------------------------
PPC460EX
PowerPC 460EX: 600MHz-1.000GHz, DDR2, PCI-Express, PCI, Gigabit Ethernet, TCP/IP Acceleration Hardware. GPIO

Product Status: Production Release


General Description
With speeds of up to 1.000 GHz, support for floating-point operations, USB 2.0, PCI-Express, SATA, Gigabit Ethernet, security, NAND Flash interfaces and low power dissipation, the PowerPC 460EX embedded processor is ideally suited to a wide range of high-performance applications, including imaging, storage (NAS), and networking.

Features
CPU Speed (frequency): 600 MHz to 1.000 GHz Performance: 2.0 DMIPS/MHz (2,000 DMIPS @ 1.000 GHz peak) 32-KB-I/32-KB-D L1 caches, and 256-KB L2/SRAM with parity protection 64KB On Chip Memory On-chip double data rate 1/2 (DDR1/2) SDRAM controller with 32/64-bit interface, up to 3.2-GB/s peak data rate and optional ECC
Support for four banks of DDR SDRAM memory of up to 2 GB each, maximum capacity of 8 GBs with CAS latencies of 2, 3, 4, 5, 6, and 7
Two PCI-Express interfaces, one with four lanes and one with one lane; 2.5-Gbit/s full duplex per lane; compliant with PCI-Express base specification 1.1; configurable as root or end point 32-bit PCI V2.2, 3.3-V interface supporting frequencies of up to 66 MHz One port SATA controller operating at up to 3.0Gb/s. Compliant with Serial ATA II specifications 5-stage FPU with 2.0 MFLOPS/MHz (SP/DP); hardware support for IEEE 754; singleprecision and double-precision operation with thirty-two 64-bit floating-point registers On-chip IPSec/SSL acceleration (optional) NAND Flash controller. Supports one to four banks of NAND Flash memory devices; direct interfacing to discrete NAND Flash devices (up to four devices) and Smart- Media Card socket (22-pins); 4-Mbyte - 256-Mbyte devices sizes supported; 512-byte +16-byte or 2-KB +64-byte device page sizes supported; DMA support allows direct, no processor-intervention block copy from NAND Flash out to SDRAM; boot-from-NAND supported 4-channel DMA - available for internal and external use
Support for memory-to-memory, peripheral-to-memory, and memory-toperipheral transfers
Scatter/gather capability
1-channel high performance DMA for internal use USB 2.0 OTG controller and USB 2.0 Host controller operating at 1.5Mbps, 12 Mbps and 480Mbps (2) Ethernet 10/100/1000-Mbit/s, full-duplex MACs with TCP/IP Acceleration Hardware, QoS, and Jumbo Frame support, supporting GMII/MII, RGMII, and SGMII interfaces. Memory access layer (MAL) provides DMA capability to both Ethernet channels Up to 100 MHz, 27-bit address bus, 32-bit data bus external bus control (EBC) interface
Support for up to 6 ROM, RAM, or slave peripheral I/O devices
Up to four UARTs (1x 8-pin, or 2x 4-pin, or 4x 2-pin, or 1x4-pin and 2x 2-pin) Two IIC (with one integrated boot strap controller) One SPI serial interface Programmable interrupt controller with 16 external inputs, 48 internal inputs Programmable timers General-purpose I/O (64) Support for JTAG board testing, JTAG debuggers, and 4xx instruction trace interface RoHS compliant (lead-free) version available

_________________
Future Acube and MOS supporter, fi di good, nothing fi di unprofessionals. Learn it harder way!

@vox

advantages:

1, Faster clock speeds
2, Level 2 cache
3, DDR2 support (cheaper memory, not really a performance increase and may be slower due to higher latencies)
4, PCI express support.

Disadvantages:

1, Would require a huge investment in a whole new motherboard design and would probably cost more for the end use to purchase.
2, would need Hyperion to port OS4.x to another cpu (not a huge task but would need money to be spent)
3, Money
4, Money
5, There may not be any other market to sell it to and the amiga community is not a sustainable market. (speculating that Acube have another market to sell the SAM440 to and thats what made it possible)

_________________

Place it in disadvantages:
6) When 460 version will get to the market it will be as old, as 440ep is now. It's to late for 460ex. We need now to go a one step further: to the FreeScale 86xx. But since Acube is in a cooperation with AMCC, it's even better to think about it's future CPU: Titan 2+GHz. Assuming it hasn’t been cancelled already. It should appear once 32nm tech will be available.

Last edited by Radov on 25-Jun-2009 at 12:21 PM.
Last edited by Radov on 25-Jun-2009 at 12:20 PM.

_________________

IMHO the next version of AmigaOS should run on x86! It is cheap, available and faster than that Freescale crap. And it doesn't matter if you emulate the 68k on a PPC or an x86. Except that the x86 will be faster. Even an emulated PPC would be faster than the real thing. And did I mention that you can get it everywhere? And that it is cheap?

@DrZarkov

Stop this endless x86 discussion. The yield is in the PPC-Direction and the Train just passed it long time ago.

Amiga OS on x86 is AROS and thats totally buggy and unusuable and no real Amigafan see it as real alternative.

_________________
I´m back in 2023 on Classic Amiga with my A1200/Blizzard1230IB@50Mhz, 32MB RAM, AmigaOS3.2 and ROMs, Indivision AGA MK3, Author of Amiga-Flipclock (OS4)

@fingus

Why thank you! So I'm not a "real" Amiga fan? I do indeed see AROS as an alternative, but mostly because it's on x86. It's got a potential future unlike the rest because they are stuck on platforms that'll never be up to date.

@Radov

Quote:

It's to late for 460ex

I agree yes its toooOO late for that.......

_________________
http://aros-exec.org/

http://serk118.blogspot.com/

@DrZarkov

Quote:

AmigaOS should run on x86! It is cheap, available and faster than that Freescale crap

x86 is todays market..and i do agree with you matey..

_________________
http://aros-exec.org/

http://serk118.blogspot.com/

@LupusGrey



Quote:

I do indeed see AROS as an alternative, but mostly because it's on x86. It's got a potential future unlike the rest because they are stuck on platforms that'll never be up to date.

Hmm, then XBOX360, Wii, PS3, Nintendo DSi, PSP Slim did the whole thing wrong, because they USE PPC-Based technology

_________________
I´m back in 2023 on Classic Amiga with my A1200/Blizzard1230IB@50Mhz, 32MB RAM, AmigaOS3.2 and ROMs, Indivision AGA MK3, Author of Amiga-Flipclock (OS4)

Also I think that PPC460 might be too little improvement.
If baby steps need to be taken via 460, then be it.
Otherwise, let's skip/jump to the next PPC SOC generation (or start using available Cell accelerators).

(or/and develop AmigaPPC virtualbox for some x86 board in the meanwhile, intel Core2 miniITX would be nice&cool)

Last edited by KimmoK on 25-Jun-2009 at 02:48 PM.
Last edited by KimmoK on 25-Jun-2009 at 02:47 PM.

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?

@fingus

I prefer being unreal instead of fanatic.

Aros brilliant coders are making it less buggy for each day that goes by.

_________________
AmigaOS or MorphOS on x86 would sell orders of magnitude more than the current,
hardware-intensive solutions. And they'd go faster.-- D.Haynie

@fingus

Quote:

Stop this endless x86 discussion

I like fast cheap x86 not dead ppc and you can keep your ppc board thats your way of been an amigan and real amiga is 68k amigas not ppc or x86 but todays live market is x86 and i dont want to run amigaos4 on slow ppc boards..


Quote:

Amiga OS on x86 is AROS and thats totally buggy

Thats not true and it runs lovely and faster than those ppc amigaos 4.0 but its just not 100% there yet so give them a bit more time..


let me clear last one more thingabout x86..

Hyperion Entertainment got all amiga community in one direction which is ppc market where they can make a bit of money the money they lost by amiga inc by getting all of us to buy slow ppc boards and port some those linux games coz its easy to port and they are good at porting games thats what they used to do and right now they are building bit more users so they can also sell those port`ed games to ppc amigaos so they got income out of it and we have been played again just like amiga inc..

goodluck to all amigans (there is a market for as but few people trying to pull as to direction where they can make money and dont care about the actual OS or the community)

i my self will be running my amiga under my winuae(thank you so much tony for the winuae) and develop for 68k amiga`s only till we find the very right direction for amiga os.

Last edited by serk118 on 25-Jun-2009 at 03:20 PM.

_________________
http://aros-exec.org/

http://serk118.blogspot.com/

@Radov

Quote:

When 460 version will get to the market it will be as old, as 440ep is now. It's to late for 460ex.

Normally I'd agree with you but the AmigaWorld is lucky to have ANY hardware so ANY development of hardware to a more powerful CPU is a good thing. For OS4.x to continue the hardware development needs to continue too! x86 is not an option right now so get over it!

_________________
"Art challenges technology. Technology inspires the art."
John Lasseter, Co-Founder of Pixar Animation Studios

@Radov

Double Post - blasted back button... grrrHHH!!!

Last edited by BigD on 25-Jun-2009 at 03:57 PM.

_________________
"Art challenges technology. Technology inspires the art."
John Lasseter, Co-Founder of Pixar Animation Studios

Quote:

fingus wrote: Hmm, then XBOX360, Wii, PS3, Nintendo DSi, PSP Slim did the whole thing wrong, because they USE PPC-Based technology

DS and DSi use two ARM CPUs, and PSP/PSP slim use MIPS - nothing to do with PowerPC at all.

Wii needed to be compatible with GameCube so for them it made sense. Xbox360 and PS3 both did the whole thing wrong, that is correct.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

Nobody is forced to buy expensive PPC-Hardware. There´s nothing wrong that Hyperion and Acube earn Money with it.

Be happy with what we have. Remember what we have before the so called "expensive" Boards for Years: Nothing or overpriced A1 with no warranty. And keep in mind that companies like acube didn´t have amiga market on main-focus.

_________________
I´m back in 2023 on Classic Amiga with my A1200/Blizzard1230IB@50Mhz, 32MB RAM, AmigaOS3.2 and ROMs, Indivision AGA MK3, Author of Amiga-Flipclock (OS4)

@fingus
Quote:

fingus wrote: @DrZarkov Stop this endless x86 discussion. The yield is in the PPC-Direction and the Train just passed it long time ago. Amiga OS on x86 is AROS and thats totally buggy and unusuable and no real Amigafan see it as real alternative.

Too bad ppc is mostly dead on the desktop. There isnt much hardware for that cpu platform and the few hardware that exist is insanely overpriced.

I am a proud owner of a SAM440ep Flex 800 MHz, but I also have my own "iMica" Clone, running Aros (in the incarnation of Icaros). Currently I prefer the SAM, it runs smoother, and it seems faster, if I compare AmigaOS 4.1 to Icaros. And of course there is more software available, even without emulation of classic hardware. Hyperion did very good work. The SAM is nice, it is silent and it uses not much power, so it it is very friendly to the environment. But the Intel Atom hardware is more than twice as fast, you see the difference if you run Linux at both machines. It is also very silent, it also uses only few power, but the motherboard including CPU and graphic on board costs only 60 EUR! How do you want to beat that? Aros is not ready yet, agreed. But if you would make a version of AmigaOS 4.x for that Intel Atom board, how will a user see the difference? I don't care about the architecture of the CPU, if it is cheaper, faster, and more easy to write software for it, because you'll find a lot of ready developement tools and information? There is no logical reason against x86, it's all emotional. Do you remember the time when Motorola was not cool at all? Do you know why Chuck Peddles left Motorola and founded MOS? Without MOS and Chuck Peddles there would have never been a C 64 or an Apple II. The 68000 was the fastest 16 bit chip in 1979, when Jay Miner thought the first time about making a fast flight simulator, so he choose that CPU. That was 30 years ago! Is that the only reason why we should stay with lame PPCs? Because Jay Miner used a chip made by Motorola in the mid 1980s and Motorola invented the PPC? They do not even produce it any more themselves! Staying with PPC is just sentimental, but not sensible.

BTW: As a "true" Amigan you will of course agree that the Amiga 3000 was the best Amiga ever. If you own an Amiga 3000, open it, and look for the manufacturers of the chips inside. There are several made by Intel!

Quote:

Nobody is forced to buy expensive PPC-Hardware. There´s nothing wrong that Hyperion and Acube earn Money with it.

What's wrong is that it limits everything:

- going PPC limits the speed (if you forget about 360&Cell CPUs that cannot be bought in small quantities without paying a *huge* price ;)), main PPC cpus are targetted at embbed market and too slow for modern desktop use
- going PPC limits the price (PPC boards are much more expensive than standard x86 boards)
- going PPC limits the software (going x86 would mean more software could be available, because everything already works on x86... it wouldn't be magic, it simply would be easier/faster)

Of course, they don't force anyone... they are just shoting themeselves in the foot.

- Apple decided to ditch the classic MacOS and numerous attempts to bring it in a modern shape (and I think they threw way more ressources on it than Hyperion/ACube ever will).

- Apple decided to switch to x86 and since then, they have been selling a lot more machines than they every did

Now maybe it will take Amiga 10 more years to come to the same conclusion...

And maybe by then Apple will be switching to a new architecture ;)

Of course it will take time. Of course you loose PPC software. So what ? the market is dead right now. There's no progression, no look forward... Amiga needs a fresh breath, much like Apple needed one. And this won't come with 600Mhz boards and an OS lacking modern features...

Apple had way more resources for sure. So it may not work. And I'll tell you: so what ? Better dying while trying to make something great, new, fresh, powerful, modern,... than struggling with old stuff.

Last edited by Leo on 25-Jun-2009 at 07:20 PM.

_________________
http://www.warpdesign.fr/

@fingus
Quote:

Hmm, then XBOX360, Wii, PS3, Nintendo DSi, PSP Slim did the whole thing wrong, because they USE PPC-Based technology

The XBOX360 the PS3 and the Wii are consoles that are sold in millions. Each of them is sold in numbers bigger than they ever sold Amigas. For each console they developed specific CPUs and GPUs, they had bigger budgets to develop the consoles, than they had for the Amiga1000 and that was more than they had for the PegasosII (still the best PPC motherboard for the Amiga market ). The console builders do nothing wrong, but they have a market the PPC-Amiga never had.

A market of 1000-2000 users is to small for own hardware, we know it is to small for own applications and it might be to small for an own OS ..

Quote:

Nobody is forced to buy expensive PPC-Hardware. There´s nothing wrong that Hyperion and Acube earn Money with it.

Nobody is forced to buy an licensed AmigaOS. If all AOS4 owners would have invested 30% of the money, they invested for AOS4 into AROS, we could have an perfect A(miga)ROS, without extra costs for hardware development and without royalty for Amiga Inc.. And this A(miga)ROS would run on fast x86 boards.

---

So in 2 years, when development is going on, AROS will have more users than AmigaOS4.

_________________
AmigaONE. Haha. Just because you can put label on it does not make it Amiga.

I borrowed this comments from here (#27 & #28):
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=38873&forum=2&start=20&order=0