On 14-01-07 08:53 PM, Billie Walsh wrote:
On 01/07/2014 07:09 PM, Larry Stotler wrote:
On Tue, Jan 7, 2014 at 6:46 PM, Billie Walsh <bilwalsh@swbell.net> wrote:
It really depends on what your doing. I edit a lot of television shows so a Quad with a minimum of 12 gig is not to bad. It can be done with a dual core and four gigs if you got loads of time, but anything less that a quad is no fun. Well sure, in that case throwing more hardware at it makes sense. What I meant was day to day tasks like web browsing, word proc, or email My, one of many, hobby(ies) is doing genealogy and genealogy/history related websites. At any given time I might have two instances of Firefox, regular and Nightly, running with fifteen to twenty tabs on each, Gimp with a couple pictures open, and eight or ten other applications open. And that's an easy day.
Those are my "day to day" tasks. *<]:oD
It is precisely the definition of "day to day" task that is critical in this branch of this discussion. Recently, I spend my time developing web applications, and the ideal target is to have my api respond to the client in under a second, though the average, including the round trip to web services my api uses, is more like 6 seconds. While an order of magnitude slower than the ideal, it remains acceptable as my competition typically averages about 20 seconds. I need not explain to most here what factors impinge on improving the responsiveness of a website, but suffice it to say, neither older machines nor entry level current machines will perform adequately IN THIS CONTEXT. I also have programs that handle batch processing of accounting data, and these take a couple days to finish. Again, these are programs you want to run on the fastest machines you can afford. In other contracts, I dealt with risk assessment models related to environmental management. In that area, it is not unusual for a useful model in that discipline to require half a million lines of highly optimized C++ code or more. And then, after many hours of compilation, the model itself may require anywhere from a day or two to a couple weeks, depending on the details of the problem. The day to day tasks of the scientists, engineers and programmers require working with such applications; and their time isn't inexpensive. They are not the kind of staff you want sitting on their thumbs waiting for their computers to finish their task. On the other hand, I have developed embedded software, in what seems like another life, and with the sort of embedded software I worked on, a tiny footprint for the OS and application code is a priority, without sacrificing capability. And on yet another hand, if one has a number of admin assistants, each using only the usual office software, the speed with which they can do their jobs is dependant on the capability of their hardware and software. Making them use older hardware, and less capable software, can slow them down significantly, making them less efficient. I'll leave it to the accounting types out there to compute the cost/benefit tradeoffs between newer, faster, hardware vs a 5% improvement in admin efficiency (it would be as idiotic to put a super computer on each admin's desk as it would be to put an old 8080 PC running DOS there). I am sure there are people out there actually doing that kind of calculation to determine when to replace older computers for businesses of different sizes. The costs associated with staff waiting on their computers can be significant. As an aside, here, I would point ot that to simply describe tasks as "web browsing, word proc, or email" is misleading. The way I browse the web now is very different, demanding a lot more of my system, than I did even 5 years ago and neither the hardware nor the software I used back then would be able to handle what I do now. Similarly, while I do not do a lot of word processing (known, when I was young, as writing), what I do now is significantly different from what I did years ago. I do more with my word processor now, just because I can, and that is just because the software supports it and the hardware is able to run it. And I would point out that what is true of me in my limited writing, is even more true of, say, admin assistants, who are able to do more, more impressively, because the software they're using supports their efforts and creativity. But, in a sense, all this discusson of day to day tasks, and the contexts in which each of us find ourselves, is beside the point. This is open source software, and a portion of the community decided it was interested in developing support for RasPi, and went ahead and did it (or are continuing to do it - I haven't followed that much). If it is true that more effort is being put into supporting RasPi than older computers, that is probably a reflection of the numbers of people who are interested in each. It seems to me that if the question is honest, one is interested in understanding why there are so many more people interested in RasPI than old PCs. If it is not so honest, it carries an implication that those who are more interested in RasPI than old PCs are somehow wrong; and that smells of arrogance. The question of what people are interested in is a value judgement, and I see no basis for passing judgement on value judgements, any more than I would accept an adverse judgement on my taste for rib-eye steak or than I would pass adverse judgement on those who mysteriously prefer salmon. I hate salmon, but if you love it, you can have any salmon allocated to me. ;-) just my $0.05 worth (since the government of Canada has done away with the penny ;-) Cheers ted -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org