RAMBUS vindicated.

[Terra]

From the forntpage of [H]ardOPC.
Rambus wins $306 million from Hynix

Rambus has beaten Hynix Semiconductor in its patent infringement case, with a court ordering its rival to pay $306.5 million in damages.

A jury found that Hynix had infringed 10 of Rambus' patents in a verdict announced on Monday in U.S. district court in San Jose, Calif. The award covers compensation for Hynix sales in the United States between June 2000 and the end of 2005, according to a Rambus statement. Hynix may also have to pay interest on the amount awarded in the verdict.

In the next phase of the trial, starting this summer, Rambus will seek an injunction barring Hynix from selling DDR, DDR2 and SDRAM. Rather than face an injunction, defendants often agree to pay additional amounts to the plaintiffs through a license.

I for 1 am happy now.
I got RAMBUS with my P4, and I like it.
And I got quite upset when I saw all the marketing FUD being spread agaisnt RAMBUS.
And I was sad when Intel annoced they where moving away from RDRAM again.
I was sooooo looking forward to QUAD-channel RDRAM PC1200 ( 9.6G) back in 2003.
Hell when I got my RDRAM PC500 the best DDR was single channel DDR2100...
Instead I saw DDR-RAM makers break the patents and make a better technology dissapear from the market....

Well, payback is here...
Hynix was the first.
Now the rest will follow.
And please don't scream about this a bad for consumers.
You have gotten you cheap "illegal" DDR-RAM..now pay the price for it
DDR-compaines broke the law, and are now paying the price for being cheap...
an not just 1 patent.
Or 2
Or 3
Or 4
Or 5
Or 6
Or 7
Or 8
Or 9
But 10!!!

I guees it is easy to make cheap RAM, when you steal other people's idea's....

Terra - Pats his old RDRAM rig...Damn I wish RDRAM would appear on the PC market again...

 

[(cf)Eclipse]

interesting. keep in mind two things:
1. the only people who really seem to care about rambus these days are battling it out in the "conroe debunked" thread
2. while it is good for rambus, keep in mind that hynix ALREADY has a 44.29% tariff on the chips they export to the US. add a $306.5 million fine, and i wouldn't be surprised if they have some serious troubles down the road.

this is really bad for the consumers. hynix is a big player in the market, despite the tariffs, they have some of the cheaper IC's around. now because rambus played their hand after everyone started using the patents, they have to pay up.. which is gonna mean higher pricing and less innovation until they swallow the hit.

it's gonna be especially interesting to see what happens with micron and samsung.. the other 2 of the top 3.

you might wanna read this before you go too pro-rambus
http://investor.rambus.com/downloads/2006-02-21 Micron EDVA RICO complaint.pdf

 

[GilmourD]

Man... I feel like Rambus just rammed me in the butt with no lube...

Look at it this way... Even Intel realized that they had to jump ship. That, to me, says a lot.

 

[Redbeard]

Man... I feel like Rambus just rammed me in the butt with no lube...

Look at it this way... Even Intel realized that they had to jump ship. That, to me, says a lot.

Well, when your competitors sell better products at half the price, you have to change something.

Hence Conroe....heh.

 

[Terra]

interesting. keep in mind two things:
1. the only people who really seem to care about rambus these days are battling it out in the "conroe debunked" thread

I bet you Hynix, Micron and Samsun care about RAMBUS right nw

2. while it is good for rambus, keep in mind that hynix ALREADY has a 44.29% tariff on the chips they export to the US. add a $306.5 million fine, and i wouldn't be surprised if they have some serious troubles down the road.

Ten they sholdn't have broken the law...

this is really bad for the consumers. hynix is a big player in the market, despite the tariffs, they have some of the cheaper IC's around. now because rambus played their hand after everyone started using the patents, they have to pay up.. which is gonna mean higher pricing and less innovation until they swallow the hit.

Again, you DDR guys now have to pay the price for the cheap "illigeal" DDR RAM..the price DDR manufactos didn't want to pay...seems only fair to me.
No awards for breaking the law.

it's gonna be especially interesting to see what happens with micron and samsung.. the other 2 of the top 3.

Yup...but my RDRAM is Samsung...wonder if that will factr in somehow...from RAMBUS point of view.

you might wanna read this before you go too pro-rambus
http://investor.rambus.com/downloads/2006-02-21 Micron EDVA RICO complaint.pdf

From what I can read, it's a counter-lawsuit from Micron against RAMBUS?
Didn't Hynix also try that?

Terra - And 10 patents just don't go away

 

[Terra]

Man... I feel like Rambus just rammed me in the butt with no lube...

Look at it this way... Even Intel realized that they had to jump ship. That, to me, says a lot.

It had nothing to do with preformance.
It was due to all the FUD, and people screaming about DDR was cheaper...
Sure is was cheaper...it was "illegal"...

Terra - RAMBUS didn't do you anything....it wasn't them that spread FUD and broke the law

 

[Terra]

Well, when your competitors sell better products at half the price, you have to change something.

Hence Conroe....heh.

Please don't Make this into Intel vs. AMD.
But since you think DDR RAM is better, why did quad RDAM have 9.6GB/s already in 2003...Compare that to DDR RAM at the time..and the imagine if RDRAM had evolved like DDR did into DDR2...

Terra - I wonder if this would have done anything for the Netburst architechture though...if RDRAM had gotten the chance...

 

[Redbeard]

Please don't Make this into Intel vs. AMD.
But since you think DDR RAM is better, why did quad RDAM have 9.6GB/s already in 2003...Compare that to DDR RAM at the time..and the imagine if RDRAM had evolved like DDR did into DDR2...

Terra - I wonder if this would have done anything for the Netburst architechture though...if RDRAM had gotten the chance...

I'm not trying to make it into Intel vs. AMD. I happen to love (and use) both manufacturer's products. What I'm saying is that Netburst can't scale as well as Intel had hoped, and for an equivalent price right now for most uses, the Athlon64 processor outperforms the Intel Pentium 4/Pentium D processor. Intel realized this years ago, no doubt, which is why they switched architectures towards Conroe, which is very exciting. That's all I'm saying.

As for comparing RDRAM to DDR, there are some major caveats.

RDRAM had higher latency, higher manufacturing costs, and was much more difficult to test than SDRAM or DDR. Not to mention the fact that RAMBUS wasn't the easiest company in the world to work with.

The other problem is that when it came out, the Pentium 3 architecture showed no significant performance improvement because it did not require immense amounts of memory bandwidth. As a result, people got it in their heads that RDRAM was not worth the extra money (and the extra money was, and still is, significant, with modules costing sometimes 2x or more as much as the competition). When the P4 Willamette came out, however, the PC800 RDRAM stuff was very effective. The problem was that people already shied away from RDRAM because AMD's DDR performance looked so good on the Socket A platform and DDR was significantly cheaper.

Not to mention the fact that a few significant miss-steps by Intel (socket 423, being beaten to 1.0 GHz, losing the performance crown at equivalent clock speeds) made consumers and enthusiasts wary.

Lastly, the Licensing fees and overall cost/performance ratio didn't quite meet what the resellers felt was fair, so they moved to a non-proprietary standard. This wasn't really theft, in fact, a number of RAMBUS's patents were openly discussed by a variety of manufacturers at JEDEC as "a possible open standard in the future" and RAMBUS went out and got patents on property they themselves "borrowed" from other companies, then turned around and sued those companies for producing those products.

To be perfectly fair, the reason RAMBUS failed wasn't because the technology was bad; it wasn't. The reason RAMBUS failed is because they burned JEDEC and they hassled the other DRAM manufacturers, and as a result they got slapped down again and again and have been forced to spend the last 4-5 years in court. If they had tried to work within the system, there'd have been none of this, but instead, they attempted to push a proprietary standard that stole intellectual property from others and passed it off as their own, which didn't sit well.

If they came back with an open standard part that outperformed DDR2 and DDR3, I'd be all for it. I'm all for performance, I could care less who makes it.

 

[Terra]

I'm not trying to make it into Intel vs. AMD. I happen to love (and use) both manufacturer's products. What I'm saying is that Netburst can't scale as well as Intel had hoped, and for an equivalent price right now for most uses, the Athlon64 processor outperforms the Intel Pentium 4/Pentium D processor. Intel realized this years ago, no doubt, which is why they switched architectures towards Conroe, which is very exciting. That's all I'm saying.

Okay

As for comparing RDRAM to DDR, there are some major caveats.

RDRAM had higher latency, higher manufacturing costs, and was much more difficult to test than SDRAM or DDR. Not to mention the fact that RAMBUS wasn't the easiest company in the world to work with.

Again, when I got my P4, the alternative was singlechannel DDR2100..and that would cripple my CPU.
I went for preformance

The other problem is that when it came out, the Pentium 3 architecture showed no significant performance improvement because it did not require immense amounts of memory bandwidth. As a result, people got it in their heads that RDRAM was not worth the extra money (and the extra money was, and still is, significant, with modules costing sometimes 2x or more as much as the competition). When the P4 Willamette came out, however, the PC800 RDRAM stuff was very effective. The problem was that people already shied away from RDRAM because AMD's DDR performance looked so good on the Socket A platform and DDR was significantly cheaper.

I can't seem to remember that "good" preformance?
http://enthusiast.hardocp.com/article.html?art=Mjc1LDMsLDIw
AMD VIAK266A:

This was when the Northwood 2.4Ghz came out...a while later then the Williamette you are talking about.
(Talk about a trip down memory lane *L*)

Not to mention the fact that a few significant miss-steps by Intel (socket 423, being beaten to 1.0 GHz, losing the performance crown at equivalent clock speeds) made consumers and enthusiasts wary.

And now it became Intel vs AMD...not RDRAM vs. DDR RAM....

Lastly, the Licensing fees and overall cost/performance ratio didn't quite meet what the resellers felt was fair, so they moved to a non-proprietary standard. This wasn't really theft, in fact, a number of RAMBUS's patents were openly discussed by a variety of manufacturers at JEDEC as "a possible open standard in the future" and RAMBUS went out and got patents on property they themselves "borrowed" from other companies, then turned around and sued those companies for producing those products.

They way I saw it, was that people turned away for the better thcnology, but I do remember the debacle over patents.
I guees we will never know the full thruth since both side accused the other of almost everything you could imagine...

To be perfectly fair, the reason RAMBUS failed wasn't because the technology was bad; it wasn't.

Agreed...I psent some time watching the "DDR-horde" slowy moving up to my RAM pseeds..and wondered why they where so happy with less preformance.
Hell, back in those day the "rumour" was that RDRAM was slow...
I had to show benches all the time to disprove the false information/rumours...and then people did 2 thing.
They either entered a state of denial.
Or look like they had fallen from the moon...

The reason RAMBUS failed is because they burned JEDEC and they hassled the other DRAM manufacturers, and as a result they got slapped down again and again and have been forced to spend the last 4-5 years in court. If they had tried to work within the system, there'd have been none of this, but instead, they attempted to push a proprietary standard that stole intellectual property from others and passed it off as their own, which didn't sit well.

If they came back with an open standard part that outperformed DDR2 and DDR3, I'd be all for it. I'm all for performance, I could care less who makes it.

Same here, but I still think RDRAM had some very strong points:
Signaling Level (RSL)
The no "loss" from going from READ to WRITE in RDRAM
Row/Column adress
The I/O driver
High bandwith on small mumber of pins
Upgrade granularity
Dissipation of I/O power
Bandwidth per watt
And then I like the heatspreader

Terra...

 

[SatinSpiral]

i wanted to buy rambus stock seven months ago at $11.50 but my goddamn wife said no. I knew they were a bargain then, and figured they would hit $50 within a year or two. here i am, validated, and still poor. oh well, at least i bought it at 11.50 in my virtualstockexchange portfolio to be able to go back and show people who doubted me. What clued me in was when they won a prior suit against them, and were slated to be used in the PS3. Oh well, at least INTC is still trading for cheap.... maybe i can convince the wife to let me buy based upon the damning rambus evidence i have against her. could have quadrupled our money if she had only listened to me.

 

[Terra]

i wanted to buy rambus stock seven months ago at $11.50 but my goddamn wife said no. I knew they were a bargain then, and figured they would hit $50 within a year or two. here i am, validated, and still poor. oh well, at least i bought it at 11.50 in my virtualstockexchange portfolio to be able to go back and show people who doubted me. What clued me in was when they won a prior suit against them, and were slated to be used in the PS3. Oh well, at least INTC is still trading for cheap.... maybe i can convince the wife to let me buy based upon the damning rambus evidence i have against her. could have quadrupled our money if she had only listened to me.

And the lesson here:
Don't ask your wife next time *L*

Terra...

 

[SulSeeker]

Well... this just f***ing bites...

I applied at the Hynix plant here in town, and even if they did have a position for me, I don't think they can pay me as a potential worker after a lawsuit like this, and if the upcoming injuncition against sales goes through... sounds like alot of workers aren't going to be working. I'm not really concerned about DDR vs Rambus debate. All I know is that it sucks that I might not be working for a good company that makes money, employes people (and friends of mine,) and makes a good product.

That's my reason for not liking Rambus right now.

 

[(cf)Eclipse]

Again, when I got my P4, the alternative was singlechannel DDR2100..and that would cripple my CPU.
I went for preformance

you did what was right at the time. these days, rdram doesn't stand a chance. if the tech really was better, and the performance was there vs dual channel ddr, people would have used it.

 

[Terra]

you did what was right at the time. these days, rdram doesn't stand a chance. if the tech really was better, and the performance was there vs dual channel ddr, people would have used it.

Quad-channel RDRAM at 9.6GB/s is far form "obsolete"...even today...and that was 3 years ago SiS did that...

And the preformance was there....people just thought with their collective *biiiip*'s

Terra...

 

[noobtech]

Rambus is our friend.

 

[CCUABIDExORxDIE]

Rambus is our friend.

i thought that too...until i saw the price

 

[(cf)Eclipse]

Quad-channel RDRAM at 9.6GB/s is far form "obsolete"...even today...and that was 3 years ago SiS did that...

And the preformance was there....people just thought with their collective *biiiip*'s

Terra...

not obsolete, no, but it can't compete with dual channel ddr2 at 500mhz (16gb/s theoretical) and ddr2 is still scaling upwards. even ddr can easily reach a theoretical bandwidth of 9.6gb/s while undoubtably maintaining far lower overall latency.

however, if you're only gonna be concerned about bandwidth and not latency, what about video cards? 3 years ago was.. around the 9800/5800 era? those cards easily pushed >20gb/s...

 

[GilmourD]

It had nothing to do with preformance.
It was due to all the FUD, and people screaming about DDR was cheaper...
Sure is was cheaper...it was "illegal"...

Terra - RAMBUS didn't do you anything....it wasn't them that spread FUD and broke the law

Wah wah wah... FUD. That's what everybody cries when they're wrong. I worked for a system builder at that time, and probably put a couple hundred systems of each type. I owned an Intel rig at that point. My personal experiences with the Athlon and Intel/Rambus made me jump ship and go AMD.

FYI... Last I checked, signing an agreement to be part of a public standards committee, and then patenting such things behind said committee's back is breaking the law.

 

[GilmourD]

not obsolete, no, but it can't compete with dual channel ddr2 at 500mhz (16gb/s theoretical) and ddr2 is still scaling upwards. even ddr can easily reach a theoretical bandwidth of 9.6gb/s while undoubtably maintaining far lower overall latency.

however, if you're only gonna be concerned about bandwidth and not latency, what about video cards? 3 years ago was.. around the 9800/5800 era? those cards easily pushed >20gb/s...

Hi, Terra. I see you've met (cf)Eclipse. Prepare for more raw data, proof, an analyzations to be presented to you than you can read before Conroe is released.

 

[(cf)Eclipse]

Hi, Terra. I see you've met (cf)Eclipse. Prepare for more raw data, proof, an analyzations to be presented to you than you can read before Conroe is released.

time to run home and fire up my p3 rig with pc800 rdram in it. until then, no benchies

 

[noobtech]

i thought that too...until i saw the price

The prices are steap now, but if it became as mainstream like DDR or DDR2 then it will eventually decrease.

 

[(cf)Eclipse]

The prices are steap now, but if it became as mainstream like DDR or DDR2 then it will eventually decrease.

intel tried to make it mainstream. it wasn't cheap then either.

 

[GilmourD]

intel tried to make it mainstream. it wasn't cheap then either.

Intel went as far as selling it as a bundle with their motherboards with a discount, and it STILL wasn't cheap!

 

[Redbeard]

Okay


I can't seem to remember that "good" preformance?
http://enthusiast.hardocp.com/article.html?art=Mjc1LDMsLDIw
AMD VIAK266A:

that shows exactly what I'm talking about. AMD processors were cheaper, and the memory bandwidth performance was close. Though RDRAM was faster, it wasn't faster/$, and since most games are limited more by CPU and Video card than by memory bandwidth and gaming pushes the enthusiast hardware segment, well...I'm preaching to the choir, everyone here knows this stuff.

We never really got too deep into RAMBUS at Corsair, at the time most of the gaming community was into AMD and DDR was much cheaper and in much higher demand.

RAMBUS's main problem, like I said earlier, has always been questionable business practices and not necessarily the technology.

 

[Terra]

not obsolete, no, but it can't compete with dual channel ddr2 at 500mhz (16gb/s theoretical) and ddr2 is still scaling upwards. even ddr can easily reach a theoretical bandwidth of 9.6gb/s while undoubtably maintaining far lower overall latency.

however, if you're only gonna be concerned about bandwidth and not latency, what about video cards? 3 years ago was.. around the 9800/5800 era? those cards easily pushed >20gb/s...

Again, I am talking about 3 year old technology.
Add 3 years of "evolution"..and you get the picture
RDRAM PC1200 was AFAIK better latency wise than DDR PC2700

Terra...

 

[Terra]

Hi, Terra. I see you've met (cf)Eclipse. Prepare for more raw data, proof, an analyzations to be presented to you than you can read before Conroe is released.

Yeah?
Point being?
It's no the first time we post in the same thread, and surely not the last?

Terra - Sorry if you don't feel you can contribuate...

 

[Terra]

time to run home and fire up my p3 rig with pc800 rdram in it. until then, no benchies

PIII?
I am talking P4 northwood and forward

Terra...

 

[Terra]

that shows exactly what I'm talking about. AMD processors were cheaper, and the memory bandwidth performance was close. Though RDRAM was faster, it wasn't faster/$, and since most games are limited more by CPU and Video card than by memory bandwidth and gaming pushes the enthusiast hardware segment, well...I'm preaching to the choir, everyone here knows this stuff.

Like I stated, I didn't like wate my FSB, so I got RDRAM?

We never really got too deep into RAMBUS at Corsair, at the time most of the gaming community was into AMD and DDR was much cheaper and in much higher demand.

I clearly remember that.
AMD-people where hating RDRAM like hell...even if it was for P4's...look at the what was being said in those days.
Like I said...a lot of FUD about the preformance

RAMBUS's main problem, like I said earlier, has always been questionable business practices and not necessarily the technology.

Agreed, IMHO their thecnolgy was better than DDR-RAM...

Terra...

 

[GilmourD]

Terra - Sorry if you don't feel you can contribuate...

Apparently you haven't read my other posts... Typical... Contextual selectiveness.

The crux of the matter is... Rambus gained its footing illegally. It's screwing the entire industry because its greedy.

Why hasn't JEDEC as a whole come in and dropped the hammer on Rambus?

 

[Terra]

Apparently you haven't read my other posts... Typical... Contextual selectiveness.

Sorry that I don't have a personal "vendatta" against you and read every single post you write.
I prefeer to answer to what people say/write to me, not what they say/do else where.
And I honestly can't remember seeing your nick before...or understand what your posts elsewhere have to do with your post directed at me here?

The crux of the matter is... Rambus gained its footing illegally. It's screwing the entire industry because its greedy.

Why hasn't JEDEC as a whole come in and dropped the hammer on Rambus?

If they did it illegally, why didn't the verdict reflect that?
And you rquestion is a good one.
If RAMBUS is the criminal, indeed why no action from JEDEC?

Terra...

 

[GilmourD]

Sorry that I don't have a personal "vendatta" against you and read every single post you write.
I prefeer to answer to what people say/write to me, not what they say/do else where.
And I honestly can't remember seeing your nick before...or understand what your posts elsewhere have to do with your post directed at me here?

Uh... It was in this thread? Wow...

If they did it illegally, why didn't the verdict reflect that?
And you rquestion is a good one.
If RAMBUS is the criminal, indeed why no action from JEDEC?

Terra...

Well, let's see if we get an answer on that one.

 

[(cf)Eclipse]

Again, I am talking about 3 year old technology.
Add 3 years of "evolution"..and you get the picture
RDRAM PC1200 was AFAIK better latency wise than DDR PC2700

Terra...

hmm, well let's see. i bought my kingston hyper-x 3 years ago. it's rated to 5ns cycle time, aka pc3200, and 2-2-2-6 timings... and then you speak of rdram pc1200.. which barely made it to market.

as for latency, i'm honestly not sure who's the winner on that. however, i know that pc1200's 'rated latency' is 32ns. i'm not entirely sure, but i think that is just the theoretical time from requesting data to the data coming out. for ddr, that would be cycle time * ( tcl + trcd + trp), which puts the bh-5 chips in my hyper-x rated to 30ns.
pretty similar if you ask me, and yes, that would mean that pc2700 is slower. but comparing pc1200 and pc2700 three years ago is like comparing conroe to K8. one's new and hasn't really hit the market while the other has been around for a while.

btw, let's see the pc1200 get a 40% oc at rated timings...
(which puts Trac to.. 21ns )

PIII?
I am talking P4 northwood and forward

Terra...

no, you're talking about rdram. cpu is blind to the ram used.

 

[Terra]

hmm, well let's see. i bought my kingston hyper-x 3 years ago. it's rated to 5ns cycle time, aka pc3200, and 2-2-2-6 timings... and then you speak of rdram pc1200.. which barely made it to market.

PC3200=DDR400
Data Rate=400MHz
FSB=200MHz
Peak Bandwith=3.2 GB/s(giving 6.4GB/s i dualchannel config)
Even if we go:
PC4200=DDR533
Data Rate=533MHz
FSB=266MHz
Peak Banwith=4.2 GB/s(giving 8.4GB/s in dual channel config)

Quad channel RDRAM(Wich DID come on the market ~3years ago):
PC1200=RIMM 4800
Data Rate=1200MHz
FSB= 600MHz
Peak Bandwith=2.4 GB/s (giving 9.6GB/s in quad channel config)

In my math that still gives RDRAM an advantage of 1.2GB/s...3 years ago...

as for latency, i'm honestly not sure who's the winner on that. however, i know that pc1200's 'rated latency' is 32ns. i'm not entirely sure, but i think that is just the theoretical time from requesting data to the data coming out. for ddr, that would be cycle time * ( tcl + trcd + trp), which puts the bh-5 chips in my hyper-x rated to 30ns.
pretty similar if you ask me, and yes, that would mean that pc2700 is slower. but comparing pc1200 and pc2700 three years ago is like comparing conroe to K8. one's new and hasn't really hit the market while the other has been around for a while.

Both PC1200 and PC2700/PC3200 appeared around the same time, I fail to se your logic, sorry?
And PC800 PC and PC2100 where "playmates" ~4 years ago...

btw, let's see the pc1200 get a 40% oc at rated timings...
(which puts Trac to.. 21ns )

OC'ed compared to what?
PC3200?
3200x1.4x2=9.96GB/s...wich means that dualchannel PC3200 needs and OC of 40% to exceed quad channel RDRAM...that isn't OC'ed.
I don't know how well RDRAM OC's though, never tried OC'ing my rig(only my recent GFX card has OC at times(most just to fiddle around with it)...as I find my system componets at stock settings macht each other quite well

no, you're talking about rdram. cpu is blind to the ram used.

But a P-III is unable to utilize the fullbandwith of RDRAM PC800 as the P-III is FSB 100Mhz or am I mistaken?

Terra...

 

[Least]

Wow, these arguments aren't circular at all

I would've killed for a p4 with rdram several years ago. too bad it cost too much.

 

[GilmourD]

Wow, these arguments aren't circular at all

I would've killed for a p4 with rdram several years ago. too bad it cost too much.

Like I had said earlier in the thread (but Terra conveniently ignored), I was working for a system builder around the time of RDRAM's peak. I built plenty of RDRAM based Intel machines and probably an equal number of Athlons. I was, at that time, rockin' an Intel rig. My experiences with those machines brought me to the AMD camp.

Before I get accused of being susceptible of FUD, one must realize that I'm in possession of some really different opinions than most, which have nothing to do with this thread, but usually get the same amount of FUD thrown around. I do things based upon experience and research, not marketing and FUD. Marketing and FUD gets you to spend money on something that you might not want. I want what I want, and everybody else can go screw if you want me to get something different.

My biggest issue with this case... In the end, the consumer gets the screw. Rambus got greedy by patenting part of a standard that it suggested to a standards committee and then waited until it was in fact the undisputable standard before saying anything. If that's not sheisty, I don't know what is.

 

[placeboFx]

I built a Rambus P4 system and was quite happy with it. The best part was the fact that all the downtalk made it so cheap toward the end of its cycle.

 

[Terra]

Like I had said earlier in the thread (but Terra conveniently ignored), I was working for a system builder around the time of RDRAM's peak. I built plenty of RDRAM based Intel machines and probably an equal number of Athlons. I was, at that time, rockin' an Intel rig. My experiences with those machines brought me to the AMD camp.

I would like to hear about these experiences?
Not that I don't belive you, I always like inputs.
And I like them to be a bit more more specific, not "vauge", sorry

Terra...

 

[GilmourD]

I would like to hear about these experiences?
Not that I don't belive you, I always like inputs.
And I like them to be a bit more more specific, not "vauge", sorry

Terra...

Well, what would you like to know? Remember, this is a few years ago, so I'm not going to be all like "On machine #347...". I built a ton of machines, and the Intel/RDRAM ones never lived up to the hype. Hell, PIII/SDR machines were spanking some of them in certain areas at the time. Gaming on AMD was a nicer experience back then. It was enough to convince me that my needs as a gamer and general computer user were better served by AMD. Same performance for much less money. Oh, sure, you can benchmark and maybe get a few more megs per second of bandwidth... But it's like the difference between a 1 caret diamond at market value and a 1.1 caret diamond that's overpriced due to some hype -- they're both nice fucking rocks, but I'd be buying the 1 caret.

 

[|CR|Constantine]

From the forntpage of [H]ardOPC.

And I got quite upset when I saw all the marketing FUD being spread agaisnt RAMBUS.

Well, payback is here...
Hynix was the first.
Now the rest will follow.

This is by far the most off handed response I have ever witnessed on a BB. After almost 6 years you still have not seen the big picture of what really happened.

In basic terms, Intel worked in conjunction with Rambus to patent an open source technology (which has never been done in the industry) in order to force manufacturers to pay royalties. Intel owned a fair portion of Rambus Inc and had the strength to literally corner the market and profit from it.

Intel was by the far the majority player at the time and now they were looking into getting a cut from all RDRAM sales. They lied to the public about RDRAM's performance in order to convince everyone that this was the best out there. They justified there moves against the open source standard by claiming that RAMBUS was far ahead of JDEC and that it had to do it.

It only happened once but Tom from Tom's Hardware did something truly remarkable. He posted a detailed scientific results that clearly showed RDRAM was indeed worse than regular DRAM.Yes RDRAM had tremendous bandwith when you gave a command one at a time, due to its massive latency's (compounded when you used more modules) programs like games that have massive amounts of commands a second did quite poorly. Due to Tom and his findings, He literally blew the lid off of Intel's/ Rambus charade which sent their stock from $900.00+ dollars a share to the low hundreds within days!

Because of Intel and Rambus, manufacturers like Samsung spent incredible amounts retooling there facilities to make this new product and spent a cut of its profits to pay for royalty's. Instead of Intel (the lion of the industry) continuing to work on pioneering the DRAM spec with JDEC they brought a new player to the table that hurt the natural progression to keep moving things forward.

Intel buyers like myself who spent butt loads on the newest chipset for RDRAM and for it's memory found a product that saw literally no gains and at times slowdows in intensive command sequence programs like games.

It was a terrible and black moment in the computing industry and I for one am happier than sh1t that AMD and JDEC/DRAM came to save the day. This whole sorrid affair was all about greed and nothing to do with technology.

Rambus and for a part Intel should be burned to the ground for what they tried to pull.

The fact that your happy about this decision speaks volumes about you.

 

[Terra]

This is by far the most off handed response I have ever witnessed on a BB. After almost 6 years you still have not seen the big picture of what really happened.

And:

The fact that your happy about this decision speaks volumes about you.

I wonder if you are so correct, that RAMBUS won the courtcase then?
If you words where true, they should have lost, not won...

Terra - Unless I have misunderstoood something in the verdict?