Talk:ARM9

CPU Speed?

so do all these processors run at 69mhz then? i heard that's what speed the DS ones runs at.

I've also read the Nokia N-Gage has the same chip. But the wiki article for that console says, "Its main CPU was an ARM Integrated (ARMI) compatible chip (ARM4T architecture) running at 104 MHz". 86.135.163.139 (talk) 12:55, 17 April 2008 (UTC)

Heck no. See chip vendor docs for what they support. There's no particular lower limit; AT91 chips come up running off the 32768 Hz "slow" clock, and can go slower. The upper limit is partially constrained by the particular design from ARM, partially by the chip vendor and the process it uses. Most ARM9 vendors seem content to synthesize and validate designs with about 200 MHz upper limits, but I've seen them clocked at over 400 MHz.

Of course, low power designs won't push higher clock speeds; and memory speeds matter. On-chip SRAM, now you're probably talking 50-80 MHz. External SDRAM, maybe up to 200 MHz or so. DDR2 gives an incentive to go over 300 MHz. 69.226.236.175 (talk) 03:32, 26 August 2009 (UTC)

literal error?

In the article the term "Harvard architecture" means "Harvard architecture cached" as the term which is used in the manual. right?

—Preceding unsigned comment added by 208.184.6.120 (talk) 11:59, 18 November 2008 (UTC) 

I think so too, but am not sure. Somebody add some more info please? --194.197.235.58 (talk) 07:14, 5 April 2009 (UTC)

Normally one talks about Harvard architecture since cache is just an optimization ... and not all ARM9 chips use cache. 69.226.236.175 (talk) 04:11, 26 August 2009 (UTC)

Why the ARM9 article redirects to the ARM9E article?

I was wondering why the article ARM9 redirects to this article (ARM9E) and not the other way around? The ARM Information Center calls both families "ARM9". The "E" in the name seems to inform that the processors are DSP-enhanced. But not all ARM9 processors have the DSP instructions. For example, the ARM920T and ARM922T processors are not DSP-enhanced.--Henrique Camargo (talk) 01:27, 28 December 2008 (UTC)

Don't know what you're talking about; maybe it got fixed? 9E redirects to 9.

69.226.236.175 (talk) 03:34, 26 August 2009 (UTC)

Yep, it got fixed. =) --Henrique Camargo

Is ARM9 von Neumann or Harvard?

I added a note to the lead paragraph pointing out this apparent conflict. An expert is needed to clarify this. David spector (talk) 20:34, 17 December 2009 (UTC)

• Could be packaged as either! Best think of it as Modified Harvard, as written. Note that the WP article on that topic was pretty crappy; it's better now but still nasty. Whoever was writing that kept talking about Harvard not Modified Harvard, too. The ARM ref you pointed out was a bit confusing, but that seems to be because it was addressed to folk integrating unspecified ARM cores onto chips ... where the silicon designer has many integration options, including "throw away the split i/d advantage, make like von Neumann" as well as the recommended "use that split i/d advantage". Go look at for example the ARM9E-S core specification and you will observe that indeed there are separate instruction and data busses ... 32 bits address and data each, total 128 bits (plus control) in parallel. Clearly Harvard ... at least, until you integrate it with a DRAM controller and let the core access both code and data managed by that controller. And on top of that, that ARM text was poorly written. I updated the WP page to talk a bit more (clearly) about that. --69.226.238.251 (talk) 13:48, 22 December 2009 (UTC)

ISA / Arch Referencess

• http://www.scss.tcd.ie/~waldroj/3d1/QRC0001H_rvct_v2.1_arm.pdf
• http://simplemachines.it/doc/arm_inst.pdf
• http://netwinder.oregonstate.edu/pub/netwinder/docs/arm/ARM7500FEvB_3.pdf
• http://www.heyrick.co.uk/assembler/qfinder.html

Some of these should probably be eventually incoroporated/referenced in the article? —Preceding unsigned comment added by Parallelized (talk • contribs) 10:26, 21 November 2010 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified 2 external links on ARM9. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20150411211220/http://www.arm.com/products/processors/cortex-a/arm-mpcore-sample-code.php to http://www.arm.com/products/processors/cortex-a/arm-mpcore-sample-code.php
• Added archive https://web.archive.org/web/20131022223108/http://digi.com/products/wireless-wired-embedded-solutions/software-microprocessors-accessories/systems-on-chip/arm-microprocessors/ to http://www.digi.com/products/wireless-wired-embedded-solutions/software-microprocessors-accessories/systems-on-chip/arm-microprocessors/

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at {{Sourcecheck}}).

Y An editor has reviewed this edit and fixed any errors that were found.

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 08:53, 1 October 2016 (UTC)

Actually, no, you didn't just modify 2 external links. You modified one external link and left the other one alone. The one you changed - the first one - worked, but I changed the reference to use {{cite web}}. I fixed the second one - the one you didn't change - to go to a new location, the link for which still works (and points to a page that specifically discusses the two chips in question). Guy Harris (talk) 17:00, 1 October 2016 (UTC)