processor upgrade for inspiron 9300

$90 for an extra .5GHz

I personally say no and save the money for something else, like a faster drive. You'd see better performance improvement there.

cheers ...

Sounds like a good plan.

cheers ...

OK let me say something some might not agree with. For a perceivable user experience you need about 20%. You are at about 30% so above? Unless budget is the issue I would get a new system? But that is just me. But yes you could get a minor improvement.

Get new?

You can be a good working in financial powerpack

20%, 30% sound much larger than .5 But yes, we do agree on "minor" improvement

cheers ...

I have been struggling with this issue as well...

Just can't justify the $90 for a 780/2.26ghz chip, but have been weighing the pros/cons between two other lower price options:

Current prices on eBay:
780/2.26ghz/533 CPU ~$90-100
770/2.13ghz/533 CPU ~$40-45
735/1.7ghz/400 CPU ~$20
745/1.8ghz/400 CPU ~$35

So, for $40 I could simply drop the 770 into the socket and have an easy boost from 1.73ghz to 2.13ghz (23% boost).

Orrrrrrrr...
For only half that price ($20), I could pinmod a 735 chip and have 31% boost up to 2.26ghz. But with the possibility that it might not work at all, or that it might need some voltage mods as well.


-$20 more for assured compatibility and no modding.
-Or save $20 and get a slight bump in speed with a bit more hassle. (and a 20% chance it won't work at all...)

What say the gurus?

Now this opens a different opportunity. If you are willing to play, I'd go with the pin mod route first before spending the extra $40

Take it as a learning lesson fee.

cheers ...

Well, actually, I even found this listing which has 6 of the 735/1.7/400 CPUs available for a BIN price of only $13.50 or best offer with free shipping!

http://cgi.ebay.com/Intel-Pentium-M-...item563967a8cb

And there are a few completed listings for 735s that finished below $10 including shipping...

------
As far as doing the pinmodding, I am not afraid of that. Just not sure that I wanted to risk the cash, as I am on a fixed income as a disabled vet. Every sheckle counts...

I am a mechanical engineer with some experience with modding desktops.
I even hotwired 5-1/4" floppy drives onto the motherboards of the old IBM PS-2s back when Big Blue dropped support for that format with the introduction of that line.
I also overclocked all of my old PDA's all the way back to the old Casiopea line by connecting/cutting traces on the circuit board.

I have mini pliers, forceps, and magnifying eyeshades and all of that...

But for the price, yea, looks like I aught to give it a try...

Like I said, go for the pin mode first before committing to buy the "upgrade". Worse case you will be busy doing something else other than on the net for a few days

Good luck.

cheers ...

Like qhn has stated the most cost effective method is going the "pin mod" route. However be aware of these caveats, you're talking about chips that have been available now for over 5 years, many of the "good" as in "overclockable" chips are now gone in the wild. Your odds of finding one is much less than it was a few years ago. But since the cost is so low, if it doesn't work you could always recoup your investment by reselling again, but it will most definitely take time to do so.

When I originally purchased my i9300 it came w/ a 1.7ghz Pentium-m. I upgraded mine to a 780 / 2.26ghz and have not regretted it, even though at the time it set me back nearly $300. I was able to successfully "undervolt" it in both Linux and Windows so it ran with even longer battery life than it did with the slower processor.

I read from your other post that you plan on keeping this one for a while as a "Desktop" machine and buying a smaller lighter weight one for taking on the road. This being the case I don't think you should be scared of making a few improvements. I still use my i9300 as my one and only take everywhere box and it serves me well.

I see you have an X-300 graphic card. If you're interested, I've got a Geforce go 6800 "vanilla" with only a few months worth of use, that I'd part with cheaply. (I did the 7800 GTX upgrade when my machine was just a few months old). It's a drop-in replacement that you could do while your machine is apart doing the CPU upgrade. I know it works, because when my screen started getting horizontal lines a few months ago I switched the cards out thinking my 7800 was the culprit. But it wasn't, it was my screen which I've since replaced.

I was thinking of selling it (the 6800 Go) on eBay since I no longer have any use for it and it's only going to diminish in value. I only have kept it around this long because of the overheating and failure issues so many were having with their 7800s. But mine has been fine, runs cool and never given me an inkling of problem, so I'm now willing to part with the 6800. Inbox me if you're interested.

I hope this helps,
Ciao

The thought of a drop-in swap with the possibility of being able to actually undervolt the CPU is certainly attractive...

And I just found someone who has 5 "Engineering Sample" Pentium-M 780 2.26/533 CPUs listed for just $75 or best offer.
I am giving serious thought to throwing him a lowball offer of $50 to see if he bites since they have been listed for well over a week now with no purchases or offers.
Seller has 100% feedback with over 375 seller reviews in the last year...

To my mind, it is hard to justify risking too high of a percentage of the cost of a "sure thing" swap on a "maybe" it will work pinmod...
If I am risking 1/3 to 1/2 the cost of the sure thing on the one that "might" work, why risk the extra exposure/loss?
If it didn't pinmod for me, I don't know if I could resell it in good faith.

What do you all think?
A Pentium-M 780 2.26 for $50-75 versus tossing out $15-20 on a CPU which will maybe pinmod to the same speed...

The potential exposure is losing $20 if it doesn't pinmod, then having to go out and buy another CPU.
If I could get them to accept a $50 lowball bid on the 780, the exposure risk of trying a 735 CPU first would be 40%.
20/50=.4

The exposure risk of trying a 735 versus a drop-in 770 is 50%. But with just slightly less speed out of the 770 @2.13ghz.
(There is one seller who accepted an offer of $40 for a 770 CPU and has 6 more for sale.)

Sorry if it sounds like I am overthinking all of this, but for us military pensioners, every dollar counts!
They have already announced that they are not giving us any cost of living increase, (and probably not next year either) even though fuel/food/rent/etc. still keep getting more expensive.
(They think the value of private homes dropping offsets all of that. Like guys on a military disability pension can afford to buy a house?!?!?!)

I am trying to squeeze the most performance out of this old 9300 that makes sense so I can to keep it viable for a while longer. And I would just hate to toss money away in the process which could have been applied directly to the project rather than in a failed attempt.
$20 is more than my daily food budget!

What has been the actual sucess rate with pinmodding the 735 (and 745) CPU's in the 9300 chassis???
What is the likely sucess rate with the chips that are still floating around out there today???

Maybe I am just too tired and need to get a good night's sleep...

Thanks again to all for your insights.
I think that I am just to close to this issue and too wrapped up in not wasting any money on the upgrades while still getting the most bang-per-buck.
I just start going round in circles in my mind...

If the pinmod works, then I have saved X amount.
BUT, if the pinmod fails, then I have lost X amount...
However, how will I know if the pinmod works without trying?
Would I be willing to sell the CPU if it fails to pinmod, knowing that the buyer is likely trying to pinmod it?
But if the pinmod works, then I have saved X amount......

GHHHAAAAA!!!!!!

Will SOMEONE please make up my mind for me?!?!?!

I am suddenly reminded of the sign I used to have up in my office when I used to work as an engineer:
"There comes a time in every project when you have to shoot the engineer, and start production!"


Goodnight Gracie...

Go for the sure thing then, in your case, and re-arrange certain other aspects of expenses (in short term) to cover the extra investment

Honestly, since it is hard to confirm for you the success rate of pin modding - it simply varies from one instance to another, from one machine to another.

cheers ...

Again I agree with qhn. Which is why I chimed in with my post. A sure thing for 50-75bucks is hard to beat. And you WILL notice a difference, as it will be snappier. Remember to keep your drive defragged, and while you've got the thing apart clean it out well and apply a good quality grease.

Now for the "undervolting" part, follow the guide in the "Notebook Guide" forum at the top of the board. I used it and it worked for me, but of course your results may vary. Especially with an "engineering sample" CPU. The reason I say this is that these CPUs are usually early production samples and sometimes don't overclock or undervolt as well as those chips produced later on. But they usually perform as advertised, especially when from reputable resellers.

Good luck!
Ciao

I have an Inspiron 9300 2.0 GHz 2G RAM NVIDIA GeForce 6800 Go

Any suggested upgrades worth the $$?

I personally would leave the comp alone or maybe a faster and larger hard drive?

cheers ...