Pokertracker load time

[Blarg]

[ QUOTE ]
256 megs is MORE THAN ENOUGH to run windows XP and poker tracker with any feasibly sized database. sure if you want to do that with 10 other programs runnings and your nightly de-frag at the same, it wont be enough... but to do JUST what you need to do at one time, it is more than enough.

If you have 1GB or RAM and your OS is constantly doing RAM swaps to disk while you JUST surf the web, you have problems

[/ QUOTE ]

Baloney. Your technical expertise has completely outstripped your common sense, and undoubtedly your actual experience too. God knows why; you must just like to argue. That is so full of crap I don't even know where to begin, and it doesn't fluster you a bit to come out with it.

If you were addressing me, I newer said that with a gig of RAM I constantly write to my hard disk. I got a gig so that would NOT happen. I will say that at 512 megs of RAM I got tons of slowdowns and sometimes even crashes when web surfing.

That all said, 256 megs of RAM is CRAP. My databases are pretty small. The problem is, you first exaggerate enormously by saying people need to be running a defrag and 10 other programs on the one hand to make it seem absurd that they could have slowdowns with normal usage, and secondly you compound that with the absurd statement that 256 RAM is fine for normal usage.

WINDOWS CAN EAT 128 megs of RAM all by itself. That leaves you with as little as 128 megs to run ALL YOUR PROGRAMS. Just opening a few programs uses up good chunks of that, much less running them and opening their various files. Where does all the memory come from when you do that? From swapping to your hard disk.

I never said I constantly grind to the hard disk, but there are frequent slowdowns when people use their computers. People DO tend to have a few things going on at once. There is no such thing as JUST surfing the web. Do you close every window once you're done looking at it? Every program? I don't. Is that absurd? Hardly.

Normal usage eats plenty of RAM. If you think 256 megs of RAM is generous and enough to run a typical user's computer without repeated disk writes and notable slowdowns, your idea of what is generous is nuts.

Just because you CAN do something doesn't mean it's being done well or at an acceptable pace, making for a reasonably pleasant experience. 256 RAM is such a crappy amount of RAM for the typical user's use that it would make his online experience a misery. If you want to scrupulously open and close every window when you're done with it or think you might be done with it, run virtually nothing unless you run it in isolation, and don't do much of anything that requires hard disk usage, then maybe you can squeak by with 256 megs of RAM, but you'll have a miserable computing experience.

But most people don't want to squeak by, and they damn sure don't want to be miserable.

I sincerely doubt you operate on a 256 meg system, and I bet if you had to, you'd be very tempted to kick it through the wall. Why on earth you would recommend a piece of junk like that I have no idea, because you certainly know better and wouldn't dream of putting up with it yourself. Your chance of going anywhere and saying 256 megs of RAM is fine for today's typical home user is non-existent.

Why is a system that's utter crap such a great idea for others? No way in hell would you be caught dead using a system like that yourself. At least, if you're like 99% of other people out there.

But perhaps you're not?

[Cosimo]

[ QUOTE ]
a master in computer science has almost nothing to do with software. it is nearly all hardware related...

[/ QUOTE ]

I'm not sure that you know who you're talking to in this thread, but I suggest that you read this paper.

[edrugtrader]

sucked back in...

bottom line: the general public does not understand how computers work, or use their own computer to it's fullest potential.

the computer i am on right now is "x86 family 6 model 8 stepping 6" with 256 MB of ram. running NT and a novel client. this thing is old. after i turned off all the unneccessary stuff, and OS feature ram hogs, it is "zippy" no resonse delays at all. i play 3 tables of UB at once while i argue with you morons. no delay whatsoever.

total ram usuage with about 15 icons in the dock such as virus shield, AOL IM, 4 novel tools, MSN messenger etc, on top of the OS, and a few IE windows and total RAM usage is 107MB.

you're making up stats to prove a point that:

1) doesn't matter.
2) is wrong.

look at your numbers... block size 4K, average file size 3,000K, average fragments per file 3.

?!?!?! do you not understand that that is ideal?!?! you're "example to prove me wrong" did just the opposite. the most commonly used files only need to do a seek after 250 HITS, and on average only 3 times per FILE request.

on hard drives full of MP3s (which is usually the case when 250GB drives are full), they could be 100% fragmented, and the disk would STILL not require to be defragmented. i don't think you understand what cases defragmenting benefits, or you certainly would not have posted your data.

again, i'll try to be done with this. how about you do the same.

[edrugtrader]

[ QUOTE ]
I'm not sure that you know who you're talking to in this thread, but I suggest that you read this paper.

[/ QUOTE ]

good show. i read yours, now you're welcome to come over and read my 3 degrees anytime you're available.

i don't think YOU know who YOU are talking to.

also, my dick is huge.

[astroglide]

[ QUOTE ]
"x86 family 6 model 8 stepping 6"

[/ QUOTE ]

aka "i have no idea what cpu/core i'm using so i'll quote from the system summary"

[ QUOTE ]
running NT and a novel client

[/ QUOTE ]

do you mean novell? netware? very advanced!

i'm pretty convinced you're trolling at this point. you're yanking "facts" out of the air and i completely proved that the defrag tool is recommended to be run with a drive that has never been over 50% full. you claimed that couldn't happen. as you can see, the analysis shows that virtually the entire thing is fragged.

[astroglide]

[ QUOTE ]
now you're welcome to come over and read my 3 degrees anytime you're available

[/ QUOTE ]

are you employed?

[Richard Berg]

Look Blarg, you're the one who thinks you know how to "tweak" XP better than Microsoft's best kernel engineers yet can't make a machine browse the web acceptably with 512MB of RAM. You're the one who thinks that demand paging (an integral feature of every modern OS since the Burroughs 6500, notice) is something to be avoided. You're the one who thinks that "commit charge" is a measurement of NT's allocated virtual address space. You're the one who claims that being a gamer adds to your authority. In short, why should we believe anything you say?

[Richard Berg]

[ QUOTE ]
SSE sarcasm etc.

[/ QUOTE ]
The point -- which would be evident had you quoted in context -- is that I'm familiar with performance bottlenecks at every level of computer engineering. I know what makes a CPU fast, an OS architecture fast, and a design/algorithm fast. Most people think that low level optimizations (throwing more computation at a problem, either via more hardware or aligning the latencies inside vector ASM code or whatever) provides the best benefit, but anyone with a background in discrete math realizes that most people are simply wrong.

I admit I have not run a debugger/profiler on the apps in question. However, consider it a relatively well-informed opinion that PokerTracker is not CPU bound.

[ QUOTE ]
It's not like poker clients are getting more efficient. And it's not like new applications coming out need less RAM and less of a CPU than before.


[/ QUOTE ]
We're talking about PokerTracker, not PartyBloat v15. (Ok, that's a nitpick.) Point is, we're not asking him to cure cancer on this machine. We're asking him to render a couple bitmaps, send a few IP packets, and run occasional queries on a 100k-row database.

[ QUOTE ]

Your suggestions remind of the mechanics that tell people to keep their old cars even though the repair costs are equal to or above the yearly costs of owning a new vehicle.

[/ QUOTE ]
Not at all. I told him to spend ~$160. You can't get a new machine for that.

If he does decide to spend more money, the parts I suggested will continue to serve him well in future boxes. I just don't advocate spending good money where it's probably not needed. $700 would buy a kickass new machine, but $700 could also be the difference between his ability to move up in limits. All I want is to provide him with the facts so that he can make his own decision. Unfortunately, some people in this thread seem intent on spreading FUD and ignorance, and in our audience's interest I can't let it slide.

[edrugtrader]

ASTRO, this isn't my computer... i'm staying with a friend who is doing 6 months in a hosptial to get her masters. i'm staying in a dorm type building in the hospital and this is the only computer i have access to to get on the internet. i have no idea what the specs are and i gave you all the information i have.

[edrugtrader]

[ QUOTE ]
are you employed?

[/ QUOTE ]
i recently quit my 100k/year silicon valley job, in which i held a senior IT position to play poker full time.

for the last year i had been playing 6 hours a day and made more than i did at my "real" job. now i'm playing 10 hours a day and building a small fortune.

i know computers, i know poker. you seem to know buttsex judging by your choice of avatar, but hey... that's OK.