AMD themselves explained this, they reiterated their stance that it is an 8 core, which it is... but they settles because they didn't want the distraction.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Intel Fires Shots At AMD For False Marketing Of Boost Clocks
- Thread starter VortexA1
- Start date
AMD themselves explained this, they reiterated their stance that it is an 8 core, which it is... but they settles because they didn't want the distraction.
12.7m reasons why. its a old processor they no longer make or care about
and the payout was only 12m then made so much money off the cpu and 12m is a drop in the ocean for them. its dragged on to long now people who brought the cpu how many years ago now? will now get about 30$ each if you brought at x time and live in floridia area?but the biggest thing i really like to see and why do people bring up class actrions for is the law firms get 3+m for doing it sorry but a law firm will have m,ade a lot more then customers they repensenting what is the world coming to
Not sure, my hunch is it would have cost too much and seeing the guy that brought the law suit didn’t understand his work load or take cost into consideration I doubt he would have had the funds for AMD to recover. I’m surprised AMD didn’t TBH but it may have been looked at that the FX chips are so far in the past that AMD felt it better to just draw a line under them and move on.
You could always contact AMD for a comment on why they decided to pay out.
Last edited:
Intel will tell you most buy the K SKU not to do any OC at all... Same with the X SKU and XE.
Roman is completely correct we have tested a lot of CPUs between us and boards. The boost AMD is declaring is at "zero" workload if your lucky (often stock does not reach this) which is pretty shocking whatever way you spin it!!!
The product is generally good, this is not...
Last edited:
Permabanned
- Joined
- 28 Sep 2018
- Posts
- 0
Their sales and marketing seems to have gotten them into this mess. They wanted to tout an increase in clock speeds as consumers are numbers driven but in reality, they can't get there often are having to stretch the truth.
If the 3950x chips are heavily binned 3900x as you'd expect, their ability will be the final call on the desktop line up.
Going forward, I doubt you'll see anything out of AMD communications without a fine tooth comb parsing by their legal team.
Good chips but big PR baggage to carry forward. Intel would be outright stupid to not pounce on this.
Yeah I definitely think there's an element of this. If they advertised identical clocks to last generation the chips still would've been ~15% faster but they probably thought less informed consumers might not be convinced they're worth buying.
-----------------------------
Here's my pitch to be in charge of AMD's marketing: literally stop advertising clocks. Clock speeds are ultimately irrelevant, what is important is performance. Obviously the clock speeds can be quoted on our website or in spec sheets, etc. but it doesn't need to be in the name of the product like it is now. We're supposed to highlight our strengths, not weaknesses. For example, we'd want OcUK to list the product as:
Ryzen 5 3600 Six Core 65 W (Socket AM4) Processor - Retail
rather than the current:
Ryzen 5 3600 Six Core 4.2GHz (Socket AM4) Processor - Retail
We have a core count and efficiency advantage over Intel, so let's advertise that. We already differentiate performance with model numbers: an R5 3600 is better (somehow) to an R5 2600, this is obvious to anyone comparing the model numbers. If we want to make it even clearer, we could always call it an R5 3620 or something. "But how will people compare our products to Intel's?" By looking at prices and reviews; we certainly don't want people to compare them based on GHz! Stop playing into Intel's hands by participating in the GHz war: we will never win that battle. We can push clocks all we want, but we need to stop reinforcing the idea in consumers' minds that performance is wholly or mostly down to clock speed.
-----------------------------
They could even start using model numbers that reflect relative performance based on some benchmark like they did in the Athlon days. e.g. Ryzen 5 4xxx where xxx is some multiplier based on something (e.g. 100 = performance of an R5 1600 or something). This gets tricky though because what benchmark do you use to be fair but also demonstrate a wide range of scenarios?
Last edited:
Associate
- Joined
- 18 Jan 2008
- Posts
- 705
- Location
- UK
Indeed. And $12 million is not that much money to settle something like this, so AMD will probably be happy to put it behind them. Pushing on would be a risk because courts don't understand technology in general, and certainly not CPU architecture. Being right doesn't ensure a win. AMD might end up losing and facing a much larger bill.
Tells me that cluster multi-threading is better than simultaneous multi-threading.
FX if done now would be a good product. It was just way ahead of its time, because all apps relied heavily on single or dual-core performance...
Small cores but many of them now.
I’m pretty sure the 3900X is a 38/3700X chip and a 3600. I’m also pretty sure the problem is related to the boosted thread running on the slower chip.
Caporegime
- Joined
- 8 Jul 2003
- Posts
- 30,063
- Location
- In a house
del
@gerardfraser Did some tests with Various clock speeds, 4250Mhz gave the best overall gaming performance according to his tests.
Use DRam Clac to tune your RAM timings for its worth and prepare to be blown away....
Associate
- Joined
- 13 Sep 2010
- Posts
- 2,346
3700x on X370 Taichi.
When I first installed the CPU (BIOS 5.60) and before I made ANY changes, it was boosting to 4400 in CB r20.
I enabled PBO, tested again, saw reduced clocks (4375) so reset it back to stock, still 4375, CB scores have increased with subsequent BIOS updates but maintained 4375 max frequency.
At a guess, I'd speculate that the 4400 I saw may have been a reporting error though, similar to the 'clock stretching' seen when heavily undervolting.
In any case stability and performance seems better on BIOS 5.80 and I'm very happy with the performance bump from my 1700, hopefully it will continue to get better with the upcoming AGESA.
I do kind of see what people are saying about false claims, however it's running great and I'm not going to get my pants in a bunch over 25MHz
When I first installed the CPU (BIOS 5.60) and before I made ANY changes, it was boosting to 4400 in CB r20.
I enabled PBO, tested again, saw reduced clocks (4375) so reset it back to stock, still 4375, CB scores have increased with subsequent BIOS updates but maintained 4375 max frequency.
At a guess, I'd speculate that the 4400 I saw may have been a reporting error though, similar to the 'clock stretching' seen when heavily undervolting.
In any case stability and performance seems better on BIOS 5.80 and I'm very happy with the performance bump from my 1700, hopefully it will continue to get better with the upcoming AGESA.
I do kind of see what people are saying about false claims, however it's running great and I'm not going to get my pants in a bunch over 25MHz
Currently I can get peak boost to 4575.5 (trying for 4600 tonight) but I have seen maybe 1% performance increase over my previous max boost of 4350.