If nvidia can sell the TI4200 for $200 and still make a worthwhile profit, and it uses almost the same core as the TI4600 then how much profit do you think they are making on the TI4600? 150%?
Printable View
If nvidia can sell the TI4200 for $200 and still make a worthwhile profit, and it uses almost the same core as the TI4600 then how much profit do you think they are making on the TI4600? 150%?
I think you're a little confused. The reason why the geforce 4 ti4600 is so much more expensive is not just cuz its core is higher clocked but it uses much faster 2.8n's memory compared to the ti4200's 4n's memory. Speed doesn't come cheap you know ;). Beside the core for the ti4200 is probably just the crap cores out of the assembly lines that can't be clocked up to ti4600 levels and they just used it for the ti4200.
I wouldn't say they are the crap cores...I mean give them a little credit :) But yes the difference is the RAM.Quote:
Originally posted by Sgt_Strider
I think you're a little confused. The reason why the geforce 4 ti4600 is so much more expensive is not just cuz its core is higher clocked but it uses much faster 2.8n's memory compared to the ti4200's 4n's memory. Speed doesn't come cheap you know ;). Beside the core for the ti4200 is probably just the crap cores out of the assembly lines that can't be clocked up to ti4600 levels and they just used it for the ti4200.
Well they're the cores that aren't capable of being clocked at ti4600 speed. This kind of practice is also done by Intel and AMD, I think...If it can't be clocked at their rated speed, it'll be knock down to another level :).Quote:
Originally posted by dudewth
I wouldn't say they are the crap cores...I mean give them a little credit :) But yes the difference is the RAM.
I believe this is called binning. They test the chips at speeds and the highest ones are made into the highest speed (i.e. ti4600). That is why the ti4400 will only oc to just about ti4600 speeds and not too much higher. Those gpu's didn't really make it that good to ti4600 speeds and were made into ti4400. So basically, the ti4200 are the loser chips. However, some of the so called loser chips (1.6A rings a bell) make it to 2.4 or so and shows that these chips still have potential.
profit is not a 4 letter word. If someone thinks you can mfg and sell for 150% they have never been in the mfg business. Only a high volume product with few or no changes in product mgf cost can be sold at low mark up. It needs to have a long life and low up front mfg cost to produce any product approaching 100% mark up. It is a lot more complicated than you think. Call your local insurance agent and ask him about product lialibilty insurance just as a start. Dont worry about the EPA or OSHA or the other gov agencies that want to get involved with your mfg process. This is ignoring the other red tape involved in business and wait till you largest customer goes bankrupt. You can not even go get your product that you were never paid for. The court takes it and sells it at auction and usually the only ones that get any money is the court and the lawyers. A real piece of cake this mfg business. That said- if every thing goes well you may end up making $20 per hour or less than some of your key people. Every few years you will have a good year that makes it worth all the effort and headaches. You try to hold on to it, but it seems as if it always ends up back in the business and so it goes. When you are old and looking back you will ask yourself if it was really worth it.
YES but not for the money
You're right...it is called binning.Quote:
Originally posted by chev1
I believe this is called binning. They test the chips at speeds and the highest ones are made into the highest speed (i.e. ti4600). That is why the ti4400 will only oc to just about ti4600 speeds and not too much higher. Those gpu's didn't really make it that good to ti4600 speeds and were made into ti4400. So basically, the ti4200 are the loser chips. However, some of the so called loser chips (1.6A rings a bell) make it to 2.4 or so and shows that these chips still have potential.
Do you know why "SOME" of those 1.6A oc to 2.4ghz with no prob? Simple, when demand for them is high, Intel will just take even the higher quality core for let's say the 2.2ghz northwood and just underclock it to 1.6ghz. Companies doing that is really rare unless the demand is high such as in the case of the geforce 3 ti200 and the ti500. At first the demand was really high for the ti200 and leadtek apparently used ti500 cores and 3.8n's memory on the first batch, only later moving to slower 4n's memory.Quote:
Originally posted by chev1
I believe this is called binning. They test the chips at speeds and the highest ones are made into the highest speed (i.e. ti4600). That is why the ti4400 will only oc to just about ti4600 speeds and not too much higher. Those gpu's didn't really make it that good to ti4600 speeds and were made into ti4400. So basically, the ti4200 are the loser chips. However, some of the so called loser chips (1.6A rings a bell) make it to 2.4 or so and shows that these chips still have potential.