Short answer is No, if your main use is gaming.
No real reason to buy 2 extra cores for just gaming.
I believe you will need z390 for 9900k and would definitely want the best vrms for the power required.
No info on release date yet either
-
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.
Core 9000 series
- Thread starter orbitalwalsh
- Start date
Short answer is No, if your main use is gaming.
No real reason to buy 2 extra cores for just gaming.
I believe you will need z390 for 9900k and would definitely want the best vrms for the power required.
No info on release date yet either
These are my thoughts also. But there is another thing that I'm considering. And that is if AMD release a 2800x also. If it is the case I might go with the 2800x. And then if Ryzen 2 is a big enough jump, I'll pop one of those into my system.
Hang on, they been testing the i9 processors on z370 boards as seen in leaked benchmarks and the motherboard vendors have already officially released bios support for the i9 chips. Would be pretty crappy if intel now make consumers upgrade the motherboard for 1 particular chip.
Soldato
He's wrong, they work fine on Z370, even the BIOS update files tell you this from the motherboard manufacturers themselves.
Soldato
Can't believe nobody has picked up on this.
The wink means only one thing: details inbound. Come on @Gibbo what you got?
Was I the only one who noticed this?
2800x in the pipeline maybe?
Last edited:
Just because the i9 9900k works in the motherboard doesn't mean the vrms are upto the job... I'd buy a proper board for the chip, much like I wouldn't run a 2700x on my ch6, I mean I could, but I'd be better off using the newer x470 board as you get the better features. I'm not sure how the power phases and vrms compare between the 2 boards either.
My point was, your better off using the boards designed for the chips to make sure you get all the performance safely
My point was, your better off using the boards designed for the chips to make sure you get all the performance safely
I agree with this, I have the maximus formula x so I would assume that board has good enough vrms on it to run a i9 9900 at stock with good stability
Soldato
Soldato
Hmm winking at a post related to 10 core 2800X.... Interesting.
Soldato
I can pretty much guarantee that 90% of the Z390 boards will be identical to the Z370 equivalents with respect to the power delivery system used, and the main difference being the features that were missing from the rushed out Z370 chipset, which already feature in the likes of the H370 chipset. I am glad to see you have been paying close attention to Intel marketing materials though
On top of that if you just a gamer, dont bother with either the 2700 or 8700k chips, HTT does little for gaming and the 2600 is proven to be almost like a 2700 for gaming.
I still having a chuckle at the journalist who is struggling to understand how the 8 core intel chip beats the 6 core + 6 thread older intel chip at games.
I am disappointed tho no one commented on the link I posted, people here still struggling to understand how much HTT has been overhyped. Putting thread count ahead of per core performance e.g. even if half the threads arent using dedicated cores. The 8600k was a monster chip, beating 7700k in HTT friendly loads and decimating it in non HTT friendly loads for a much lower price, showing how bad HTT is, and people still swamped to the 8700k chip over the much cheaper 8600k.
Of the new chips the i7 9700k is the jewel, 8 core 8 threads, multiplier unlocked, but I bet people will swamp to the 9900k instead.
yes HTT has never been worth much if anything in most real world workloads.
If the numbers look poor it doesnt mean its a poor test, it is e.g. way better than cinebench which is a very unrealistic test. You can clearly see geekbench scales when new cores are added to a cpu so it does scale with more cores, its just that people have unrealistic expectations from fake (logical) cores.
Why is that odd? A 8/8 chip is superior to a 6/12 chip. Higher numbered chips should be superior.
Ouch, HTT has been a huge money spinner for intel, the 8700k £120 more than a 8600k, about a 35% price increase for nowhere near that level of performance increase, for most people a 8700k is probably single digits better than a 8600k and only in some situations.
If the 9700k is around £320 thats a pretty good deal. If they maintain the same 35% HTT tax.
They may well be binned lower, HTT creates heat, to counter that heat, lower voltage, so the HTT chips probably have favourable binning, so they can hit their speeds at better voltages so heat is managed. However otherwise they the same chip with just a bit less cache. Both are 8 core chips, HTT been a popular gimmick that is allowing intel to sell chips for inflated prices. I dont agree with the other poster who thinks the 8/8 chip has no purpose, it will have 97% the gaming/desktop performance of the 8/16 chip for 65% of the price.
Yeah now thinking about it a bit more, intel have not put this up as a 9600k, but as a 9700k, suggesting they not going to price it at the normal level of their non HTT chips. my earlier reply assumed that so I actually dont think this will be cheaper than a 8700k if the 8700k is kept on the market, if the 8700k is removed from the market then it may be cheaper, but it looks like intel may try to price the 9900k in a new price point above where the i7 chip usually is and the 9700k somewhere between that and the i5 prices. If the 9700k is 35% cheaper than a 9900k, in fact I would say anything over 20% cheaper, then its a no brainer to choose a 9700k over a 9900k, as HTT only shines in very specific workloads. Also the fact as you increase core count, the gains in the real world decrease as software wont utilise all the cores so readily. But bear in mind your looking at it from a standpoint that core count = performance, currently clock speed and IPC is still king, with core count a secondary factor for gaming. HTT is pretty much irrelevant for gaming so when I say core count I mean physical cores.
But if you comparing this to older gen intel chips like the 8600k and ryzen chips like the 2600x, then it may well be poor value, I wouldnt consider a 2700x for gaming unless it was discounted to 2600x price levels, as they both perform the same.
They done it because there is a market for it.
There is people like me who dont think paying 35% extra for HTT is worth it and if I was buying a new intel chip I would for sure be picking the 9700k.
You make it sound like that if someone thinks 8/8 is better than 6/12 they an idiot, but 8/8 is actually better than 6/12.
To be fair that is realistic, I cannot remember the last time I ran an app or a game that worked every core real hard, I dont think I have ever done, the only time thats been done on any of my systems in the decade or so I have used quad or more core cpus is stress tests and benchmarks.
If I was the intel CEO I would have monitored sales numbers following the leak of meltdown, if they held up reasonably well I would not have prioritised it. Most of spectre and meltdown will probably never be exploited much in the wild, its pretty difficult to pull off, but they were massively overhyped in the media and got people scared, my desktop PC currently has none of the patches installed, windows is patched to dec 2017. That is my confidence in it. Of course it is possible it is prioritised but just very hard to mitigate efficiently at the hardware level meaning they simply dont have it ready yet. For someone to pull off meltdown or spectre data leaks the system needs to be already compromised to some degree, in some cases as well it needs to be rooted which the media hasnt pointed out.
Last edited:
I am disappointed you didn’t see my comment saying the exact same thing.
https://forums.overclockers.co.uk/posts/32094434
I am disappointed you didn’t see my comment saying the exact same thing.
https://forums.overclockers.co.uk/posts/32094434
Agreed, I apologise
It's not just logical cores that Geekbench doesn't take much advantage of, it shows higher results for chips with half the number of cores sometimes: i7-7820X vs Threadripper 1950X, for example. The single threaded test seems to scale sensibly but the multithreaded one just doesn't. The numbers being poor doesn't necessarily mean it's a poor test but I am not sure what their multithreaded numbers translate to in the real world. Would one really expect an i7-7820X to outperform a Threadripper 1950X in a well threaded application? Cinebench at least represents a real-world workload, as do things like gaming and Office suite tests.
Serious security issues should be hyped by the media and often are not. The reason Meltdown and Spectre were (for a change) is because it was a big name like Intel and affected over a decade's worth of products. Apple and Google are the only other companies who seem to get mainstream coverage of security issues, from what I remember. People put too much stock in assuming their products are secure when very often they are far from it.
As someone who works in cyber security, I would say your opinion on the matter is naive and ill informed (see below).
No, it just needs the system to be running outdated software/OS/microcode/hardware, like your desktop for example.
Yes, some of the security flaws can also only be realistically exploited on a system the attackers have local access to. These types of exploits are not going to be used to steal your credit card information, they are actually more likely to be used by state actors. Some can be exploited remotely though.
Updating software is only useful in patching known security holes in software, patching software is like putting a second plaster on a wound, it should never be the primary means of mitigating security risk but as a backup only.
The main reason security experts always say update your system update your system all the time, and of course jumping on the spectre bandwagon is simply to not go against the trend and that its easy for end users to carry out. I dont care about saying it here because no one knows who I work for, and I wont lose any work from it, but I do know that if the rest of your system is locked down very well, then the practical risk of a spectre compromise is tiny.
I suppose another way of trying to explain this is look at banking security at bank outlets.
If you operated a "no compromise" policy on security, every bank would have 24/7 manning of security, an outer perimeter, customers walking in during opening hours would be strip searched, go through metal detectors etc. But they dont do any of that, they do whats cost effective and also assess the practical risks.
Working in IT security every possible exploit is assessed, compromises are decided if they worth it and not every single patch is blindly applied as if all else doesnt matter.
You not going to get a drive by infection that utilises meltdown or spectre because its not possible, you even said yourself local access is already required. If you get compromised via say meltdown, then your system was already compromised via another form of entry.
Now I do have meltdown mitigated on some systems I manage, I just dont have it mitigated on my desktop PC, I am unlikely to be the target of a state funded attack, and its very unlikely to be used by mass targeted automated malware. My desktop PC has many lockdowns that are not commonly used on consumer configurations, that malware will not expect. So dont mistake me for not patching meltdown as some kind of guy who doesnt care about security because you couldnt be more wrong, its just I pick and choose what mitigation's I deploy rather than blindly deploying everything thats possible. Its just my personal opinion that the performance hit (and stability problems I observed) of all these patches way out weigh's the pros of applying these patches. Also remember that the vast majority of internet connected devices for a long time, several years will not be patched due to a combination of old hardware and lazy operators. Yet with all these devices no one is targeting it and there is a reason for it.
Also if its not clear from my post I feel its irresponsible to publicly disclose possible security exploits, when there is no proven case of use, there is no issue in preparing patches or even deploying patches silently but it was no way a good idea to tell the public about this stuff. As it just has raised awareness now with the bad guys that this kind of problem exists in hardware. I feel it is particularly stupid given its a hardware fault, meaning its going to take a long time for the problem to go away in internet devices.
The main reason security experts always say update your system update your system all the time, and of course jumping on the spectre bandwagon is simply to not go against the trend and that its easy for end users to carry out. I dont care about saying it here because no one knows who I work for, and I wont lose any work from it, but I do know that if the rest of your system is locked down very well, then the practical risk of a spectre compromise is tiny.
I suppose another way of trying to explain this is look at banking security at bank outlets.
If you operated a "no compromise" policy on security, every bank would have 24/7 manning of security, an outer perimeter, customers walking in during opening hours would be strip searched, go through metal detectors etc. But they dont do any of that, they do whats cost effective and also assess the practical risks.
Working in IT security every possible exploit is assessed, compromises are decided if they worth it and not every single patch is blindly applied as if all else doesnt matter.
You not going to get a drive by infection that utilises meltdown or spectre because its not possible, you even said yourself local access is already required. If you get compromised via say meltdown, then your system was already compromised via another form of entry.
Now I do have meltdown mitigated on some systems I manage, I just dont have it mitigated on my desktop PC, I am unlikely to be the target of a state funded attack, and its very unlikely to be used by mass targeted automated malware. My desktop PC has many lockdowns that are not commonly used on consumer configurations, that malware will not expect. So dont mistake me for not patching meltdown as some kind of guy who doesnt care about security because you couldnt be more wrong, its just I pick and choose what mitigation's I deploy rather than blindly deploying everything thats possible. Its just my personal opinion that the performance hit (and stability problems I observed) of all these patches way out weigh's the pros of applying these patches. Also remember that the vast majority of internet connected devices for a long time, several years will not be patched due to a combination of old hardware and lazy operators. Yet with all these devices no one is targeting it and there is a reason for it.
Also if its not clear from my post I feel its irresponsible to publicly disclose possible security exploits, when there is no proven case of use, there is no issue in preparing patches or even deploying patches silently but it was no way a good idea to tell the public about this stuff. As it just has raised awareness now with the bad guys that this kind of problem exists in hardware. I feel it is particularly stupid given its a hardware fault, meaning its going to take a long time for the problem to go away in internet devices.
Soldato
If that post was made of bricks instead of letters, the Mexicans would be right where Trump wants them.
The trouble is, with the way Microsoft now patches Windows, it is futile to try avoid these patches. They will just get rolled into the next Windows release. And then you get 6-months of updates in a big bang. You can of course choose to never upgrade your Windows version, but then you are also going to miss out on new O/S features.