Wimminz – celebrating skank ho's everywhere

Had a few messages about the gigabyte brix / celeron 3150, and some of them contained links, and the upshot is according to all the perceived wisdom out there, it’s a totally crap CPU and worse value for money than a new i7

I dunno, for starters in my opinion it’s pointless to even compare a soldered system on chip device like the 3150 to a separate fit and remove at will high end cpu, the clock speeds and cache sizes do not explain the vast gulf between the 3150 and my own now older socket 2011 i7, so how does comparing crysis frame rates or arbitrary benchmarks help?

It’s like buying a vehicle solely on the basis of 0-60 times, a bog standard 600cc motorcycle will make every car out there with a very few exceptions look sick, but what you really wanted was something that transported 7000 kilos or something that transported 30 people or something that could pull a plough.

Your PC as a whole is too slow when some component or some peripheral either has an empty buffer waiting for data, or an overflowing buffer waiting to process data, that’s it, there is nothing else, until you get to certain use cases where machine A will render a 4k video to certain settings in 30 minutes and machine B will do the exact same job in 120 minutes, and even then only when that extra 90 minutes means the job missed the deadline, and even then you have to look at why the discrepancy, did one have 2 gig or ram and the other one have 16 gig, did one have 4 cores and the other have 16, and did the app in question have the ability to use the extra cores and memory, etc etc etc.

Provided teh (sic) buffers are neither empty nor overflowing, it doesn’t matter, it doesn’t matter if the cpu spends most of its time at 30% instead of 3%, it doesn’t matter if ram is 75% used instead if 25% used, it doesn’t matter if you have 4 cores or 16, it doesn’t matter if you have 1.5 ghz core clock or 3.0, all that matters is whether the buffers are neither empty nor overflowing, because that’s all that limits how responsive the machine is to keyboard and mouse input and screen display of same.

If a machine is bought to run MS Office and manage a wordpress site, it doesn’t matter if it sucks at playing crysis, and if the machine is bought to play crysis, it doesn’t matter that no more mundane task like the odd bit of emailing or web browsing can use more than 3% of cpu.

I have a home built NAS box based around an asus H81 and earlier 2 core celeron 1850 and 8 gigs of ram, a bunch of 6tb WD reds and drivepool,  it doesn’t matter it is allegedly weedy or underpowered, what matters is on read and write it can saturate my gigabit LAN, so it can’t get any faster…. sure, put it in an office with 20 users sharing resources and it will experience latency and people will find it no longer keeps up, but IT ISN’T IN AN OFFICE WITH 20 USERS, so why the fuck would i buy or build a machine that could cope with that??? I have uses here that you’d never get in an office of 20 machines…

But the fact is that benchmarks are a expression of a need to “rate” a thing, and then we use that “rating” to compare, so in reality the 900 lb gorilla in the room is this very fact.

If like me you have decades of experience in engineering and IT, you don’t need no fucking benchmarks to compare nuthin, even if it is a thing that you personally have no prior hands on experience with like the celeron 3150, all your other prior experience allows you to make a bloody good guess.

If you do not have the experience, then the “rating” is what you will use, and by definition, you will have no way of personally verifying that “rating”, because if you did, you’d create your own, which is what I did with the celeron 1850 NAS box, there is no linpack or geekbench calculations going on, there is MY benchmark for MY usage applications.

Which brings us to the second problem with “ratings”, not that many people would agree that Bo Derek was a “10” for me she was a 5 at best, and of course I never met her in real life so you can add or subtract at least 4 to that score, which does not mean she is not a 10 for someone else… not all people are the same, not all use cases are the same.

“Ratings” attempt to deal with this by having different ratings systems, but still, none of them are mine, and my system now rates at 7,000 on one set of scores and 11,000 on another, far from helping this issue, this just confuses it as much as me saying Bo Derek was a 5 at best.

Now we are at a place where we have many “ratings” to choose from, but since we cannot create our own because we lack the knowledge, we can’t make a valid choice between the various “ratings” either, we just added even more vagueness and unknowns to the problem.

So on the one hand we have “ratings” and on an entirely different planet a million miles away we have what that thing will actually do in the real world in different situations for different users, and as time passes those two planets get ever further apart.

So what has this got to do with anything apart from shitty weedy celeron CPU’s?

Well, it’s got everything to do with everything.

VW diesel emissions, S&P and Moody’s, currency exchange rates, stock valuations, foreign policies, social policies, legal / employment / health issues, economic policies, are “the russians” “good” or “bad” and how about the jews, or the saudis, or the yemenis, or the argentinians?

It’s all “ratings”, and the one thing you need to take away from this more than anything else in life, if that someone who buys and specs PC hardware based upon ratings by definition knows sweet fuck all about PC hardware, and you can substitute PC hardware for anything else you like, and the song remains the same.

AAPL is now AA+ on S&P 500 global rating, so is the US$….

The F35 is awsome, the iphoneX is awesome, amazon web services / cloud computing (someone else’s box) is awesome, none of these people know what the fuck they are talking about, if they did, they would not be using ratings… QED.

If you care about your life, you are far far far more likely to make a correct choice if you go down the route of “that wimminz blog bloke is a technical fucking genius” / “that wimminz blog bloke is a fucking clueless newb” than if you go down the route of listening to anything anyone who used “ratings” of any kind has to say…. because you can know for a fact that none of them have the slightest clue about the subject matter.

And no, do not give me that fucking bollocks about oh no, THEY know, they just have to package it as a rating so that THEIR customer / clients / users / readers can get the TL:DR version…. that is just giving a “rating” to the people giving the “ratings”