Talk:PowerPC 600

Untitled
I started this page as a replacement for the individual PowerPC 601, 603, 604 and 620 pages since they sorely lacked attention and was of very poor quality. I've incorporated everything from those pages into this one. Though they differ technically, they could be grouped together in a historic perspective, just as I've done with the PowerPC 400 family. Since these processors are of yeasteryear, I don't think that the technical similarities or differences are the main thing, but the historical similarities. Even though one might fill long articles with intricate details about either chip, I'm not the one who's prepared to write those articles, and this is the best I can do.. If I don't do something, I'd be surprised if anyone did anything to freshen up the old ones. As it stands now, the separate articles are in a pitiful state. Time is not our friend here, and soon there will be hard to find any information about these processors. And besides that.. they do share the name for some reason. -- Henriok 20:25, 27 December 2006 (UTC)

A bit daunting
Wow, this article covers a lot of ground. I took a look at the old ones, and you've added a lot of details. The sections in this article have more depth than the old articles they replaced. But there's still room to grow, and it's hard to make this page even longer. I wonder if some of these should be split back to their own articles? In particular the 603 series. Also projects in the "extended family" section really don't belong so much.

Since there wasn't a specific "PowerPC 600" product, maybe we should move this to "PowerPC 600 series"? Essentially it covers every processor introduced 1993-1996, which is a lot of architectural differences. Potatoswatter 07:17, 15 February 2007 (UTC)
 * I recognize your points, but I obviously don't agree that much if one can't fill sub pages significantly. For now, the historical perspective is so much greater than he focus on the individual parts. You are, however, more than welcome to be bold as I was. I will support any attempts to enhance articles about Power Architecture. -- Henriok 08:24, 15 February 2007 (UTC)

Sections → Discrete Articles
Hello. Various sections contained within this article should be split into discrete articles if for no other reason than the fact that citing is spotty at best and references are either messy or nonexistent. Quite a bit of the information presented is also wrong which underlines these issues. I have quite the collection of PowerPC documentation and own several processors encased in foam and plastic. Trollaxor (talk) 03:37, 4 December 2009 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 1 one external link on PowerPC 600. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Added archive https://web.archive.org/20090207004411/http://www-01.ibm.com:80/chips/techlib/techlib.nsf/products/PowerPC_603e_Microprocessor to http://www-01.ibm.com/chips/techlib/techlib.nsf/products/PowerPC_603e_Microprocessor

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

Cheers.—cyberbot II  Talk to my owner :Online 21:16, 18 January 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 1 one external link on PowerPC 600. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Added archive https://web.archive.org/20070930185324/http://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=MPC604E-ARCHIVED&nodeId=01z3Tw9059r4bp to http://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=MPC604E-ARCHIVED&nodeId=01z3Tw9059r4bp#

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

Cheers.—cyberbot II  Talk to my owner :Online 18:41, 20 February 2016 (UTC)


 * Found it on NXP's Web site. Guy Harris (talk) 19:17, 20 February 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 2 one external links on PowerPC 600. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Added archive http://web.archive.org/web/20110606041159/http://www-03.ibm.com/systems/p/hardware/whitepapers/power/ppc_603.html to http://www-03.ibm.com/systems/p/hardware/whitepapers/power/ppc_603.html
 * Added archive http://web.archive.org/web/20110606091441/http://www.ibm.com/chips/techlib/techlib.nsf/products/PowerPC_604e_Microprocessor to http://www.ibm.com/chips/techlib/techlib.nsf/products/PowerPC_604e_Microprocessor

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

Cheers.—cyberbot II  Talk to my owner :Online 09:16, 1 March 2016 (UTC)


 * One worked, the other didn't. I managed to find the elusive "IBM white paper" about the 603 on the archive of the IBM Microelectronics site (www.chips.ibm.com), and redid that reference. Guy Harris (talk) 20:16, 1 March 2016 (UTC)

The three 620 "explain" templates
Article indicates 620 was initially fabbed on .5µm process but 604e was fabbed on the .35µm process, suggesting the 604e came later. Was the 604e faster simply because of a better process node? Article also says 620's architecture (like FPU) was better than the 604's: "The floating point unit was also enhanced compared to the 604. With a faster fetch cycle and support for several key instruction in hardware (like sqrt) made it, combined with faster and wider data buses, more efficient than the FPU in the 604". So, was the 604e really better or just able to be clocked higher because of a process node advantage? Were the, I assume smaller, enhancements to the 604 architecture in the 604e really enough for it to leapfrog the 620? If so, why didn't the 620 get some enhancement itself, instead of being released with an old surpassed spec? But, since the article suggests the 620 hit the market sooner than the 604e, due to the node difference, I'm confused. Was the 604e the 620's primary competition once the 620 hit the market or did the 604e steal mindshare even if it wasn't released yet? Regardless, why didn't the 620 get an improvement to make it clearly better than the 604e? It seems pointless to release the 620 if it's based on an already-obsolete spec. Or, is the problem with the 620 that the decision was made to fab it on .5µm, resulting in a die that couldn't clock high enough and/or was too expensive in comparison with 604e? Are we talking more about value (cost/benefit ratio) or pure performance, when making this comparison? Also, the article says the 620 was node shrinked to .35. Was this too little, too late? If so, how, in relation to the 604e? — Preceding unsigned comment added by 107.77.195.142 (talk) 04:53, 30 July 2018 (UTC)
 * Basically, what I'm looking for is a clear explanation of why the 620 was delayed and why it didn't meet performance expectations. Why, for instance, was implementing the full PPC spec not enough to make it worthwhile as a product? — Preceding unsigned comment added by 107.77.195.142 (talk) 04:57, 30 July 2018 (UTC)