They seem to exist in their own version of the world these days, sadly seems to be the way quite a few companies are going - I see the same thing with Microsoft these days I guess it is partly due to so many "yes men" that are probably conveying a perspective of everything working fine to those above them as per the on paper concept even when in the realworld it isn't.
-
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.
New PT Data: i9-9900K is 66% Pricier While Being Just 12% Faster than 2700X at Gaming
- Thread starter Nelly
- Start date
They seem to exist in their own version of the world these days, sadly seems to be the way quite a few companies are going - I see the same thing with Microsoft these days I guess it is partly due to so many "yes men" that are probably conveying a perspective of everything working fine to those above them as per the on paper concept even when in the realworld it isn't.
Regardless of money making intentions basic scientific methodology needn't be expensive. Just ensure a level playing field. The results aren't accurate to 1 or 2% anyway. I'd say a 5% difference could go either way if testing different copies of the hardware in a different system.
I will agree with you that the results from paid testers for marketing purposes are useless. Sadly the uninformed masses believe them. It's them becomes a moral issue of do we let them buy something they've been misled about or make a fuss and tell the world? That's up to the individual. I'd like to think there are people out there who would help me out in areas I know nothing about.
Companies like this don't have the experience in any specific area like in this context say Gamers Nexus would have and most clients won't want to pay for the real cost of man hours and so on to get upto speed at that level unless it was testing for legal standards, etc. unfortunately these kind of companies usually have to take on a lot of more simplistic testing to make ends meet which often reflects poorly on them.
That said going back to:
For instance they were only doing 3 runs on most tests and picking the middle one which methodology wise has its pros and cons when used in a discriminate fashion (with more samples) but they weren't - particularly poor in this context as it can result in a big discrepancy between results.
Soldato
well thats Gospel then
Did you even bother to research yourself? Did you not think it odd that no other tech journalist has not mentioned this information...
Soldato
Must be true then SMH
Soldato
Don't get me wrong I wasn't saying they were wholly owned by Intel, merely pointing out that a major source of income is supplied by Intel, and that Principled Technologies premier product and benchmarking suite was developed for Intel, using Intel's money which they obviously don't like having to disclose, but it's there in the small print. I am pretty darn sure that if Intel paid PT to do a report for them, they told them exactly what they wanted testing and how they wanted it testing - after all and as you say PT are there to make money period.
It just occurred to me that the fact that we're discussing the precise methodology tells us that there really isn't much in it. If there was an attempt to skew the results it also says that the competition must be close.
All good news for the consumer
All good news for the consumer
Soldato
I really think that all this was to detract that the 9900K holds VERY little in performance over the previous generation. After checking the PT scores after they had been redone it not only showed the flaws in their testing but also little improvement in performance at a significantly increased price. While, yes it does perform better than the AMD products the price/performance is significantly lower than the competition.
So you shouldn't use a mode clearly labelled 'Game mode' in games unless you have tried both on and off previously and figured out which is best
What about 'Creators mode' is that as daft?
I'm getting the idea you're lazy. Any tech type worth calling one could find out in less than 10 seconds that Creator Mode is default with everything turned on. All the cores, all the threads. All the cpu power is available.
Which is kinda the point in bashing a tech company doing testing... because in huge amounts of media reviews and announcements you can find out the compromising that Game Mode does in order to get maximum compatability with programs which were made long before AMDs core count and newest CPU designs appeared.
https://community.amd.com/community/gaming/blog/2017/08/10/amd-ryzen-threadripper-for-gaming
More to the point exactly like you said they claimed to have tested with Game Mode on and off and they decided from their results to use Game Mode.
There is a very cynical way to see that since their second round of published results shows that with Game Mode off performance is vastly better. Either meaning they were grossly incompetent or grossly rigging the testing.
Ultimately unless you see the horrible performance suggesting compatibility issues with a new AMD CPU you shouldn't touch Game Mode. Read the manual and all that.
Soldato
Intel gave the order what to test and you can bet enabling game mode on all AMD CPUs was in there. The 2700X is Intel's direct competitor but their CPU comes in at nearly double the price so anything they can do to make it look **** and draw in loads of pre-orders is all good on their end.
Soldato
The question now is, was it just ryzens numbers that were fudged? I wonder if the Intel numbers were absolute best case scenario in some way?
Soldato
Of course they were. They even had awesome coolers on them but left the 2700x with it's stock one.
Deleted member 66701
D
Deleted member 66701
There are some specific, older games that still get tripped up by more than 4 cores.
Soldato
Thats true. Also many games from the 90s hate HT also. Good example the Ravenloft 1&2.
On CPUs with Hyperthreading on (hadn't tried it on the 1800X), had to turn it off (tested on 4820K, 6700K, 4930K, 6800K). On the 8600K has no issue out of the box.
(ahh yeah, I am AMD fanboy with 5 Intel CPUs over the last 5 years)
It wasn't just games. SQL Server 2003 (and early 2005 versions) couldn't understand the Hyperthreading concept - they saw 2 cores and would try to make full use of both of them...
Never had problems with hyper-threading aside from with the first Pentium 4 3GHz which had it that caused stutter in a small number of games. Some games people complained about problems with hyper threading was actually a HPET problem caused by their own tweaking or poor game programming and it was just incidental that turning off hyper thread changed the behaviour.
IIRC it was a bit more complicated than that - incorrectly configured the internal scheduler in the SQL software would often result in two processor heavy threads ending up "on the same core" so to speak when another real core was available, limiting the scheduler so it thought you had a CPU with less cores resulted in more optimal CPU utilisation.
IIRC it was a bit more complicated than that - incorrectly configured the internal scheduler in the SQL software would often result in two processor heavy threads ending up "on the same core" so to speak when another real core was available, limiting the scheduler so it thought you had a CPU with less cores resulted in more optimal CPU utilisation.
Associate
- Joined
- 13 Mar 2009
- Posts
- 704
The thing is once you get to a certain point in anything the price usually goes up exponentially for lesser returns, take bikes, you want to shave a bit of weight off your bike, easy to do up to a point then it starts getting very expensive. The fact it's faster even by 0.1% means to most it's better and price is irrelevant if you want the best regardless of what it is. I think the prices are crazy but I've still bought one safe in the knowledge that it's hardly any faster than my previous processor but it is faster even if it's only 4%. Yes the 9900k might only be let's be generous and say 4% faster than a processor half the price but that really isn't the point.
It's up to the individual to choose whether the price is one they are willing to pay.
It's up to the individual to choose whether the price is one they are willing to pay.
Last edited:
Associate
Guess most of us on here (maybe not Crinkleshoes, really should have got into Bitcoin at the same time as him!) will be on a budget of some kind, it all comes to the trade offs that we want to make. Given those results and the benchmarks I've seen I'd rather get the AMD processor and use the cash I've saved elsewhere to spend on a performance bump I'd actually notice.
I find myself in a strange position these days in that for instance I wouldn't even give dropping 1, 2 even 3 grand on an actual top end GPU a second thought if I actually needed the performance but I wouldn't touch say the 2080 for a penny over £500 on the other hand. And right now I don't really even need the performance - at 1440p with G-Sync the 1070 plays anything I bother with fine with most settings turned up.
Caporegime
They have to own their incompetence Rroff, they turned half the cores off on the 2700X, that's a noob mistake.
Shilling for Intel having been paid by them for it or horrendous incompetence... i don't know what is worse, whatever it is no one is being too harsh on them for it.
Last edited: