If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Greebe's juiced up Athlon @750 on an MSI Irongate Based M/B Marvel G200 TV with HW/DVD Daughtercard,
CDBurner, Creative DVD, two big WD Hdds, Outboard 56K modem
Parallel Port Scanner, Creative S/B AWE 64 (ISA), and a new Logitech WebCam (My first USB device)
Greebe, thanks - I went back an read the previous page (doh!). Bummer, now everything makes sense.
Hmmm, hopefully things work out well for both Matrox and 3DFX visa vi their respective suits. Clearly, there is a pattern to Nvidia's behavior in the industry.
I hope that Matrox is able to fully recover, I really have enjoyed their products over the past few years and their determined efforts to improve their drivers.
I've seen first hand how the loss of engineering staff can kill a whole product cycle (or even a whole product line!). And while I hope Matrox has bolstered salaries and incentives, it is nearly impossible to retain ambitious employees when they are offered the chance to make millions.
Personally I think some "cross-pollination" in the industry is good. But leaving a competitor severly damaged is definitely a bad thing (from a consumer perspective, and not too fun for the employees that stay on!).
Cheers,
-AJ
Small note, NVidia is allegedly guilty of violating the law (even though the proof you have access to is likely very clear on the actuallity of said offenses occuring). It that whole innocent until proven guilty thing (even if guilty!). That's why I made the lawyer comment.
Trying to figuring out what Matrox is up to is like tying to find a road that's not on the map, at night, while wearing welders googles!
I think the word would get around if it was done repeatedly in a relatively short period of time,not just once.
And just who would be the one that has more to lose the company who's on tight release dates,mostly because of the competition,and lost a qualified engineer or that same engineer who can find work elsewhere very easily and could potentially end up working for a competitor.
But on another note,this brings up a very important issue,the one concerning all these suits and countersuits about who developed what first.
Can we agree at least that all these engineers that work in all of the graphics related companies,are extremely smart people that allow us to see the increadible advances seen so far in only a few short years.
Can we also agree that each group of engineers that works in each and every company,knows exactly what it takes to make a killer video card that will ultimately be able to create photo realistic graphics some day(in terms of features).
But since it isn't possible to integrate that kind of functionality yet,not because they woudn't be able to do it mind you,but rather that it isn't possible to fit that many transistors on a single chip yet.
Since current high end chips are using about 25/30 million transistors and they're nowhere close to having all the needed features to do that(photo realism),i'm guessing(feel free to correct me)that it would take anywhere between 5 and 10 times the amount of transistors to acomplish that.
Here is where we come to the heart of the matter.
This means that obviously chip desingners have to make choices about what to support first right???.
But they may have to also add other features that work the same way as it does in your competitors card,but they developed them first.
Case in point,no one can argue that it was 3Dfx that had the first card that could do multitexturing in a single pass,even if it was a multichip solution.
Does this mean they can sue everyone else that added that feature after???(including matrox)
Here's another,nvidia made the first card with integrated T.L...does this mean they can sue everybody else that have(radion)and will have it???(hopefully 3dfx and matrox).
Then,of course we have matrox,who integrated environmental bump mapping and the ability to use 2 monitors etc...,can they sue as well if others pick up on the feature later???.
I think everyone has got the point by now and it will only get worse as chips get more complex.
Each and every maker will make certain features their priority,but they all know what it will take to build the ultimate superchip.
So what are all the lawsuits about...one word TIMING.Sounds very silly if you think about it,doesn't.
Here's a perfect example of a group of companies getting together and using their colective I.Q to make a better product.
The JEDEC,the standards comittee(sp) that discusses upcoming memory tecnologies,and is formed by the major memory companies and together decide what type of memory that ultimately end up in our PC's.
All of them give their advice and tecnical know how,without claiming a patent for it(it's part of the conditions for being in the JEDEC),it's all open standard.
Now i'm sure that i can hear you thinking(as you're reading this)that by now that this nut(me.. )has completely lost it if he thinks card makers are just gonna just work together on this an share their know how.
I say nope...not at all,just leave each other alone and the huge court fees that they save are better spent in R.D.
p.s,was a long post,but i think it was worth it,no???.
note to self...
Assumption is the mother of all f***ups....
Primary system :
P4 2.8 ghz,1 gig DDR pc 2700(kingston),Radeon 9700(stock clock),audigy platinum and scsi all the way...
System 1:
AMD 1.4 AYJHA-Y factory unlocked @ 1656 with Thermalright SK6 and 7k Delta fan
Epox 8K7A
2x256mb Micron pc-2100 DDR
an AGP port all warmed up and ready to be stuffed full of Parhelia II+
SBLIVE 5.1
Maxtor 40g 7,200 @ ATA-100
IBM 40GB 7,200 @ ATA-100
Pinnacle DV Plus firewire
3Com Hardware Modem
Teac 20/10/40 burner
Antec 350w power supply in a Colorcase 303usb Stainless
Wow, I can't believe I'm actually posting on this thread, but I also can't believe people are arguing this issue.
When you work for a company that has intellectual property that it wants to protect, you sign a contract when you accept the job. It absolutely standard procedure for every company I can think of. You generally have the opportunity to enumerate "inventions" covered by the contract that you developed prior to employment. Some allow provisions for doing work "off company time" in related fields, as long as you aren't selling it to a competitor. Nothing is for sure, though. As an example, the GNU Free Software Foundation will often not accept code from programmers at certain companies, because the code that they write, even in their spare time, is controlled by the company. You might think you are donating your code to the FSF, but your company may think otherwise.
This may seem restrictive if you aren't familiar with it. But it isn't illegal, if phrased properly. And it does come down to: you don't like the contract, don't accept the job.
High competition levels don't really enter into it, although I'm sure they have some minor effect on how restrictive the contracts are in a particular industy. I work for a software company with a fairly strict contract (no competitors for 2 years after quitting). This is in an industry where good developers come at a premium and it's typical for large percentages of people to jump ship for better jobs every year (between 25%-50%). The contract just ensures that the ship you jump to isn't competing against your old ship. I'm sure they loose potential employees because of the contract, but the other choice is just too costly. Taking trade secret to a competitor could cost them millions of dollars a year, which is likely less impact than loosing a potential employee every now and again.
And remember, most companies have these contracts, so it's not like you can say shove it and find 100 jobs without contracts...
[This message has been edited by DanK (edited 20 September 2000).]
In normal times where there would be more than enough engeneers to go around for everyone,i could see that companies can be more picky and have stricter conditions for employment.
But if were seing these kind of practices between company's,them there obviously aren't enough.
but since the demand for is increasing very fast and one of the reasons is that as chips get much more complex,companies will need way more people on the project as to be able to develop it as fast as possible.
Let's say you're looking for a job in the field,and you know that you have employment opportunities in a lot of places,but you'd prefer to work at matrox(example).
being the smart guy you are,you go for an interview,like what you see,the person who's responsable passes you the contract,you read it(including the fine print. ).
You see that clause,(not the Ip one) and give the contract back and tell the guy to shove it up where the sun doesn't shine,because you know than in 2 days you can be working in another company.
It's as simple as that.
note to self...
Assumption is the mother of all f***ups....
Primary system :
P4 2.8 ghz,1 gig DDR pc 2700(kingston),Radeon 9700(stock clock),audigy platinum and scsi all the way...
Then please explain it to us how patent infringement works?
On one hand a company goes to court because it feels it's rights have been abused,and it thinks it deserves compensation.
But if that same company has integrated features that have been developed a previous company(like the examples i quoted)and have the same function,can't the company who developed then first sue??.
Because the personal rights issues have had more than enough posts.
what i was wondering,and so far everyone has been ignoring,is the remainder of the post.
note to self...
Assumption is the mother of all f***ups....
Primary system :
P4 2.8 ghz,1 gig DDR pc 2700(kingston),Radeon 9700(stock clock),audigy platinum and scsi all the way...
The employment contracts can include whatever wording they want, in many U.S. states (most importantly CA, home of Silicon Valley), such non-compete clauses are simply unenforceable.
<TABLE BGCOLOR=Red><TR><TD><Font-weight="+1"><font COLOR=Black>The world just changed, Sep. 11, 2001</font></Font-weight></TR></TD></TABLE>
"The JEDEC,the standards comittee(sp) that discusses upcoming memory tecnologies,and is formed by the major memory companies and together decide what type of memory that ultimately end up in our PC's.
All of them give their advice and tecnical know how,without claiming a patent for it(it's part of the conditions for being in the JEDEC),it's all open standard."
And they also have been accused on numerous occasions of price fixing too.
"such non-compete clauses are simply unenforceable."
I don't know about the rest of y'all but I know personally of someone that not only lost his new job but also lost everything he owned, through a lawsuit, for violating the non-compete clause of a contract he signed with a previous employeer. I guess in all depends on the company.
Joel
Libertarian is still the way to go if we truly want a real change.
In short, anything you develop is free for other people to copy unless you patent it. After doing so, you have the right to sue any other company over that specific matter. In court you will simply have to prove that the technology that the other company is using, matches exactly (or VERY close) what your company invented and patented.
xortam: That would really depend on the company & the type of project requiring the use of a contract. If a project for instance has to do with a new cpu Intel is developing, and you have no choice but to work on it, failure to sign the contract will simply get you fired if there is no other position available that you can move to. I don't know how many people would risk that, so I'd say a lot of the times contracts are a 'small legality' instead of an employee choice.
You guys need to change the topic. This is Matrox Hardware, not LAW101.
I didn't know that they were accused of price fixing(JEDEC).
But then again that can apply to OPEC,who control over half the oil production in the world and is currently gouging us silly.
Anyways back to topic,i only gave that example just to prove a point.
But notice how no one commented on the middle part of my rather large last post,i find that rather interesting....
note to self...
Assumption is the mother of all f***ups....
Primary system :
P4 2.8 ghz,1 gig DDR pc 2700(kingston),Radeon 9700(stock clock),audigy platinum and scsi all the way...
Transistors are connected to photorealism in a very indirect way. Photorealism is possible to do through software, since it allows you to "emulate" non-existant hardware features so to speak. The problem is that no video hardware including the latest InfiniteReality3 system is fast enough to do that kind of thing in realtime. Sure transistor count will increase later on, but if you had an electronics background you'd see that speed and # of transistors aren't very connected. Size & speed are connected. Material & speed are connected, but count & speed.. is a little vague. Those extra transistors may just be some extra cache. If you're talking about a transistor increase without a change in surface area of the die, then that would be "size"
To comment on your "middle" part, someone specializing in law correct me if I'm wrong, but I think there are certain features which are an inevitable part of the chip development process that cannot be avoided, and over those, lawsuits get tricky. It's the same concept as MS suing Apple for the usage of "windows" in MacOS. Windows being the rectangular boxes with icons. Maybe multitexturing is different.. but I don't think by much.
Someone pointed out a while ago that a lot of the features current video hardware uses were in fact invented long before they got implimented in videocards. SGI alone should be able to do a mass-lawsuit on the entire graphics industry, given that they're sooo far ahead of the video industry right now. Their problem is the inneficient MIPS structure which increases the price 2x in proportion to performance.
Superfly, before making an argument, at least research what you're talking about, okay? You're asking a whole lot more questions than stating relevant facts.
Joel -- that's exactly my point, shouldn't this thread get moved into the soap box by now? :P
Comment