:: Re: [DNG] Purchasing a new computer…
Forside
Slet denne besked
Besvar denne besked
Skribent: Wim
Dato:  
Til: dng
Emne: Re: [DNG] Purchasing a new computer/laptop
2016-01-27 16:57 GMT+01:00 Simon Hobson <linux@???>:

> Wim <objectief@???> wrote:
>
> [snip]
> > * Yes, the one with the faulty SATA that doesn't like 3gbps drives - you
> try finding a drive these days that has a jumper for 1.5gbps operation :-(
> >
> > You don't need to jumper drives. I've never encountered any drive that
> doesn't work with these machines.
>
> You have been lucky then. I've tried several, and every drive I tried did
> not work (I think I must have tried about 3, different makes - including
> buying another new 750G that didn't work) - it would appear to connect etc,
> but once "in use" would cause constant errors. I did try the local genius
> bar, and they were more helpful than I expected (2nd hand machine, out of
> warranty). First off he was able to see from the service history that it
> had already had a ribbon cable replacement which I gather was the usual
> suspect. And he did run some tests - but as it wasn't an Apple drive he
> couldn't do much more.
>
>

I would take a look at that SATA cable AGAIN. These break far too often.
And when they break, they often don't break completely. Symptoms vary from
weird boot problems, to the OS going corrupt, to a general slow drive.

If by 750G, you mean the Seagate hybrid 750 GB drive, I have several users
with a 2008 Macbook Pro using these. I installed them and they all worked
on first try. Could be worse with "revision A" Macbooks. We don't get those
here in Europe. That's a US privilege. I only see them from people who
bought their machine while in the USA. And I can't recollect if I ever saw
an early 2008 rev. A MB Pro.



> In the end, I looked at the drive model, looked up the specs, and saw that
> there was a 500G version - this machine had 250 in it originally. So I kept
> an eye on eBay until an Apple branded one came up and got that - and added
> a second one in the optical drive bay.
>


There used to be a difference between Apple labeled harddisks and
"ordinary" HD's. It mattered when doing AV stuff on G4 or older. I can't
say there's any difference nowadays. Except for label and price :D


> Incidentally it was only one or two models that had this problem - I think
> mine was the first. Earlier models only supported 1.5G SATA, later models
> fixed the issue. These had a 3G SATA interface which was "flaky" if urn at
> 3G, but the drives Apple used were only 1.5G. It was only when people
> upgraded with "other" drives that the problem showed up.
> There's was fair bit of noise in the forums about it for a while.
>


I know. I do support for a living, with an estimated 70% of users on OSX.
Most of these users are AV professionals, so they can't use Linux. A few of
the web devs I support have a Debian Wheezy VM or partition. It's stil a
bit better to be able to test stuff on the real thing, as OSX has some
differences.


>
> > OSX even sends UNSAVED documents to the cloud, according to some.
>
> If you let it ;-)
> But yes, it defaults to pressuring you to linking up with your Apple
> account so it can automatically sync stuff. Most of the builtin
> applications now autosave as they go along - so it's not surprising that
> the autosave versions get synced.
>


It is amazingly hard to stop OSX from sending metadata. I don't have an
icloud account, but had to rip out several background services to get
Mavericks stable when doing audio recordings.


> Safari also defaults to sen


ding every URL you edit to Google (via Apple ?)


Not via Apple. That's the one thing I"m fairly sure about. Apple is
collecting data, but they will never share it with anyone else.

I've been using DuckDuckGo for a long while. It's good. I also use Ghostery
to block tracking. Amazingly good, if you take into account that it's made
by an advertising company.

- not just what you type, but the entire URL. It's also irritatingly stupid
> in that it will sometimes decide that your url doesn't look like a url and
> do a search or it will decide that your search term looks like a url and
> fail to load it. I'm of the school where if I type into a search box I
> expect it to search, and if I type in a url box I expect it to be treated
> as a url - and the two boxes are different things and should be separate.
>
>
> > I don't think legislation to keep hardware "open" will work. Legislators
> tend not to understand the matter and hardware manufacturers are way too
> clever bypassing laws.
>
> I agree. Something is only likely to happen "after the fact" if enough
> people complain. I suspect that the same law that got the guy (in the link
> I posted earlier) a refund and his legal costs paid might go somewhere. If
> the hardware won't run an "unapproved" OS, then it might be possible to
> argue that it's a form of tying - ie you can't use X unless you buy Y -
> which I wasn't actually aware was specifically illegal over here.
>
> Now, if people go into ${shop} and buy a PC, and specifically mention that
> you'll be needing to run other OSs on it, then it doesn't matter if the
> sales guy knows what EFI is or whether secure boot can be turned off or not
> - if it can't boot your OS then it's not fit for the purpose you stated
> when you bought it and you are entitled to a refund.
> Getting a full refund is really expensive for the retailer - they get a
> product back that they can't sell as new (and if it's mail.online they
> legally have to pay carriage both ways), so they will make a loss even if
> they eventually sell it as "grade B". So there is an opportunity to hit the
> retailers of such crap with costs and influence what they stock. We have no
> clout with manufacturers, but if (big) retailers decide not to stock things
> that cost them money then that will get noticed.
> I wouldn't want to do this to "the little guys", but IMO the big chains
> are fair game - and if they have a statistically significant quantity of
> returns then they are likely to be asking questions of the manufacturer.
>
>

Anyho, enough OT. Back to the core of the matter.