• 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.

*** AMD "Zen" thread (inc AM4/APU discussion) ***

Something I only just noticed on the Ryzen pages:
The CPU's integrated memory controller supports AM4 motherboards as well as speedy and energy-efficient DDR4-RAM in dual-channel mode with or without ECC error correction.
Click to expand...

And on The Gigabyte gaming mobo pages:
Support for ECC Un-buffered DIMM 1Rx8/2Rx8 memory modules (operate in non-ECC mode)
Click to expand...

Does that mean you can put ECC DDR4 in the board and it will work fine just with ECC disabled?

This is very interesting to me as at the moment you can only get "gaming" RAM without ECC which means gaudy heatspreaders and no low profile sticks.
 
problem is the people who bench games often pick the games they know will favour their product.amd for eg always benches their own games. sniper elite ashes of dog **** and so on. it would be nice to see gaming benchmarks of popular modern games across the board with both single player and mp done.mp benches in bf1 for eg is so different to single player that you jsut have to ignore any single player benchmarks period.
 
~>Dg<~ said:
problem is the people who bench games often pick the games they know will favour their product.amd for eg always benches their own games. sniper elite ashes of dog **** and so on. it would be nice to see gaming benchmarks of popular modern games across the board with both single player and mp done.mp benches in bf1 for eg is so different to single player that you jsut have to ignore any single player benchmarks period.
Click to expand...

If they did that we would get a fuller picture of how the hardware performs, you can get that from newer Youtube reviewers, most mainstream reviewers are useless. They just like stuff that takes 2 minutes to do and is easy to replicate, along with following Intel guidelines for reviews.
 
~>Dg<~ said:
mp benches in bf1 for eg is so different to single player that you jsut have to ignore any single player benchmarks period.
Click to expand...
This has been the problem for ages especially with Battlefield (pointed out lots of time with BF4 and Mantle), but to be fair to reviewers it would be a lot of effort to benchmark multiplayer with any kind of repeatability.
Really it would require a map with either a bunch of players doing (as close as is possible) the same moves, or a map with only scripted players but acting as if they were players. A massive undertaking for any review site.
The only other alternative way would be for a reviewer to use a real multiplayer map and run the test lots of times and take the averages. But lots of times might mean having to run 10, 15, or more runs. Multiply that by 10 CPUs or 10 GPUs and again it's a massive undertaking.
 
KompuKare said:
This has been the problem for ages especially with Battlefield (pointed out lots of time with BF4 and Mantle), but to be fair to reviewers it would be a lot of effort to benchmark multiplayer with any kind of repeatability.
Really it would require a map with either a bunch of players doing (as close as is possible) the same moves, or a map with only scripted players but acting as if they were players. A massive undertaking for any review site.
The only other alternative way would be for a reviewer to use a real multiplayer map and run the test lots of times and take the averages. But lots of times might mean having to run 10, 15, or more runs. Multiply that by 10 CPUs or 10 GPUs and again it's a massive undertaking.
Click to expand...

It doesn't have to be perfect, if you pick full servers, go to the same area's and then run the two recordings side by side with Frame counters pepole can see whats going on on screen and the resulting FPS, this is what Youtube reviewers do, the two recording are rarely the same but the viewer can clearly see which is better.

This excuse that its not valid because it difficult to replicate MP is just an excuse not to do it, because then they can't just put a bunch of numbers on a slide and say "there you go the big bar at the top is the winner"

Speaking of which..... these side by side runs are far from the same and yet its easy to see which is better, mainstream reviewers were asked to do this...... "no no no its not valid" Ryan Shrout: "we used a very overclocked 6 core Intel CPU and a throttling reference R9 290 and played 5 minutes of campaign mode BF4, we couldn't tell the difference between Mantle and DX11, Mantle is pointless"

What a ####'## ######## he ^^^^^ is!

 
I haven't had a lot of time to keep up with events other than a few youtube vids from RGT and AdoredTV but everything is looking pretty good for AMD, part of me is gutted I couldn't hold out for a system much longer but at the same time my 6700k is by no means of a disappointment (if I discount all the quality/RMA issues on a system I spent around 4K on). I don't even have to overclock the thing for anything other than benchmarking runs and a bit of playtime when I have the time, which is rare these days and the way it's looking it will be years before an upgrade is justifiable. Wallet:happy, Inner child:sad.

I have to say, though, I expected a lot more external PCI3.0 lanes from earlier material, although that could have all been guff, I thought some of them were AMD slides. I sort of half have a suspicion that they just cut out everything they were having issues with to get the CPUs to market and I sort of half expect the next gen, maybe even the APUs to look more like earlier reports.

The remarks/concerns about memory support/speeds very much remind me of the issues I had with my PII 940 and high-speed RAM when all slots were populated, which while resolvable is something I'm glad not to be dealing with, assuming there is actually an issue there.

I'm really interested in some of the features they have, though. XFR, I love things like this but I'm not convinced they will do better than manual overclocking, to date I haven't had the greatest experience with software or auto overclocking on motherboards but I'd love to be pleasantly surprised.

It could be 5 years until I upgrade this base system, so I guess I'm stuck watching from the sidelines. Which has its benefits in separating fact from fiction, I suppose. The only real thing of interest I have to look forward to is a new GPU because these MSI cards are the worst cards I have ever owned, ever. Their quality control has gone to ****, still wanna RMA the ******** but I'll save ranting about that for another time, when/if I have it.
 
Food for thought.
6900K is a power hungry monster running hot, with 140W TDP and with inability from design to get cooled efficiently.

1800X burns less power, for more performance l, having also 95W TDP (67% that of the 6900). It generates less heat and looks, due to design, that is getting cooled more efficiently.

That means the rumours around that is overclocking well could be true. In addition the LN2 test happened on the AMD event last week, where everyone had signed an NDA. AMD could have advice the guys to clock the 1800X as much as it required to beat the Intel previous record, which was with 5960X running at 6Ghz.
Reason is keep the hype and interest high enough for the main release event this week.
where we will find also what the X means on the 1700X and 1800X. (see Linus video it tells us so)
Also Lisa Su couldn't have been so optimistic to mention enthusiasts and overclocking all those times if the 1800X can't deliver.
This drip feeding of exciting news is common marketing practice.
If they had come all out last week they could have given Intel 10 days to sort out their strategy, before the chip was even on the market....

Food for thought
 
ViRuS2k said:
hmmm wonder if you can delid these chips or i wonder if there soldered 1800x and 1700x chips might be solder....
Click to expand...
There is a picture of one 1800X delided. Painful to see the state of the core after forcing the soldered heat shield off. Assuming is a 1800X.

but according to AMD they are soldered. No cheapening here....
 
MadMossy said:
Lets not their infamous "overclockers dream" comment about the Fury X 2 years ago, we all know what happened with that.

On another note as a company if they can't deliver this time they will be in serious trouble.
Click to expand...

Has any flagship device ever overclocked as poorly as the Fury/Fury X?

Lisa Su has a track record of overselling AMD, a few exaggerations here and there does wonders for AMD's stock price. ;)
 
You must log in or register to reply here.
Back
Top Bottom