AMD clubs Intel like a baby seal.

In article , Bill W
wrote:

I don't think LEDs are going to save a heck of a lot.
I don't leave lights on in rooms that are empty. But
I have been switching. I avoided the fluorescents --
They contain mercury, they die early, and the light
is ugly. I avoided LED at first because of the cost.
I prefer halogens. But now I can get LEDs relatively
cheap and they have a nice color. Hopefully they'll
last as long as the companies claim they do.

I've had a pretty high failure rate for LED's , but they're still
worth it. One benefit that most people overlook is that you can get a
LOT more light out of wattage limited fixtures. There is never any
concern about overloading circuits or fixtures. And if you live in a
hotter climate, they give off far less heat.

no failures here.

On Sat, 16 Sep 2017 18:30:06 -0400, nospam
wrote:

In article , Bill W
wrote:

I don't think LEDs are going to save a heck of a lot.
I don't leave lights on in rooms that are empty. But
I have been switching. I avoided the fluorescents --
They contain mercury, they die early, and the light
is ugly. I avoided LED at first because of the cost.
I prefer halogens. But now I can get LEDs relatively
cheap and they have a nice color. Hopefully they'll
last as long as the companies claim they do.

I've had a pretty high failure rate for LED's , but they're still
worth it. One benefit that most people overlook is that you can get a
LOT more light out of wattage limited fixtures. There is never any
concern about overloading circuits or fixtures. And if you live in a
hotter climate, they give off far less heat.

no failures here.

Odd. I've had several fail, I think mostly Cree's. I also started
buying them fairly early, so I hope they're a bit more durable these
days. It's pretty much all of my bulbs now.

I think those Cree's are still under warranty, but shipping costs more
than new bulbs these days.

On 16/09/2017 17:15, Mayayana wrote:
| Second, where are your back-up statistics? You're
| making a claim with no links or evidence.
|
|
| Couple of online tests:
|
| https://www.techspot.com/review/943-...e-desktop-cpu/

That's an ambiguous report. A quick check at Tigerdirect
for prices turns up $129 vs $169. So they're not even in
the same price range.

What I was meaning was same class of processor - given two processors of
roughly comparable performance - the Intel processor is likely more
efficient (does more work per watt) than an AMD processor. If you don't
accept that then fine - I will find you some more specific data to
support that proposition.

On price - yes, of course, Intel processors are generally more expensive
than AMD for any given benchmark. But that wasn't my original point -
benchmarks rarely take into account energy efficiency. If they did, the
TCO would go down, and some users would feel (more) ethically superior ;-)


And I didn't see any figures about
what they actually mean by power consumption. There are
no actual wattage numbers. And these are only power
figures for situations that run the CPU to its limits.

Indeed - some of the more thorough tests look at idle and a range of
scenarios. If you can't find anything to support your thesis, I'll come
back to this thread with some more data.

Meanwhile
my electricity costs last month were about $25. Most of that
is scam fees and taxes. ("Delivery charge", state and federal
taxes, etc.) If I accidentally leave my cellar light on for
a hour I've probably more than eaten up my alleged savings
from the Intel CPU, if they exist at all. Context. But I can
never be sure because I don't have even vague figures on
what the actual difference might be. If there were credible
figures saying that an AMD will use, say, 30 watt hours a month
more than Intel with normal usage, and thereby cost 7 cents
more, that would at least be evidence. But for me it wouldn't
be relevant... for obvious reasons.


Oh of course, yes, context. If I take an electric shower, that scrubs a
month of savings I made by choosing Intel. Why bother?

And that's a question you need to answer for yourself - no right or
wrong there. Why recycle, why avoid meat, and so on.

Mind, I'm still not sure that you accept that AMD processors are, on the
whole, less energy efficient than Intel.


But more importantly, there's a massive media invention
of sheer nonsense around these things. They're comparing
things like overclocking to perform a demanding task in
Powerpoint or 7-Zip. Who pushes their computer to the limit
working in Powerpoint? Who overclocks? I don't. If your
life is high-end video games and you're 16 years old and
you want to beat out all of your friends, then you save up
and study these tests and try to get the best possible CPU
to render incredibly complex scenes at the fastest possible
speed. For the rest of us this is just bunk. CPUs have been
far more powerful than necessary for many years now.



Agreed.

I need a CPU for my latest computer. I can pay one price
for AMD or significantly more for Intel. If it were not for AMD
we'd probably be paying in the thousands, and we'd need
different CPUs for 32-bit OSs and 64-bit OSs. (That was
Intel's plan before AMD thwarted them.)

* And no one with any sense buys a top CPU for a typical
computer. My current 8-core FX-8200 was one of the
cheapest available when I bought it.*


And a good buy it is - for you. If I bought an FX-8320 and used it 10
hours/day at not much more than idle, I'd be consuming 260kW of
electricity a year - at UK prices, about £40*. If I bought an i5 3570
and used it on the same basis the electricity would cost me £33. That
cost saving might not ever offset the higher purchase price, but that
isn't the sole reason for my choice - 'saving the planet' (sic) is there
too.

Also, inefficient doesn't necessarily mean waste - that extra heat
generated by an AMD chip may be useful in some situations. For a variety
of reasons it isn't useful to me.

*
http://www.anandtech.com/show/6396/the-vishera-review-amd-fx8350-fx8320-fx6300-and-fx4300-tested/6

Maybe the benchmarking reports can sometimes be
informative to identify lemons or peaches in the CPU market.
But in general they're useless for the average person. It
reminds me of the PC magazine articles of the 90s. They
got advertising money from hardware companies, so they
had to push hardware. Every time a new PC came out it
was "blazingly fast", and magically the PC that was blazingly
fast 6 months ago had become "good enough for email and
some web browsing". It was absurd. Much of Consumer
Reports is similar. They'll compare idiotic things like "ease of
cleanup" for water based paints, cooking up any old
"benchmark" to make their ratings seem significant.

As Trump says, it's fake news.


:-)

If you do require granular, per user, evidence that is so tightly
contrived that it could produce (and predict) accuracy to within a few
points before you would even consider a discussion on this topic, then
yes, I concede, you are correct, and I have made a bad choice of processor.

But hopefully we both know that that is impossible, and a reasonable
prediction is possible from the data available. Whether we treat all the
variables with equal weight is another matter.

--
Cheers, Rob

On 18/09/2017 09:50, Paul Carmichael wrote:
El 16/09/17 a las 20:51, Mayayana escribió:

Â*Â* I don't think LEDs are going to save a heck of a lot.
I don't leave lights on in rooms that are empty. But
I have been switching. I avoided the fluorescents --
They contain mercury, they die early, and the light
is ugly. I avoided LED at first because of the cost.
I prefer halogens. But now I can get LEDs relatively
cheap and they have a nice color. Hopefully they'll
last as long as the companies claim they do.



They don't. HTH. My house is full of them and I buy new ones every few
weeks.


Not sure how useful anecdote is, but I've been using LEDs throughout for
some years. Most of them from the pound shop, and a few on 12 hours each
day. Not a single failure yet. And because of my rather careless use
(lighting unused areas for example) they have saved me quite a few
hundreds of pounds.

--
Cheers, Rob