D
Deleted member 157276
Guest
I'm creating this thread to remind people of being cautious of overhyping unreleased products, as it seems like it has become a very bad tendency in this forum. Although Zen 2 was a very impressive and positive launch, the performance was substantially below what many overly-optimistic users had predicted on various tech communities over the last year. On Techpowerup’s forum it can be summed up best by this thread:
www.techpowerup.com
In May, 2018, several users made stupendous predictions about Zen 2's performance improvements, of which the predominant one was that of a user who claimed that a 20% IPC + 20% frequency + 50% core increase was well within range:
These opinions were of course disapproved by several users, as they displayed a severe lack of understanding of how technology works. Process node shrinks that do theoretically allow for these performance increases almost never really ever do so in practice (as virtually all cases of CPUs with node shrinks over the years have shown); the advertised improvements from node shrinks are always exaggarated, and never represent real-life cases. Nevertheless, the above user would have none of it, and also had his opinions vehemently defended by other participants:
The discussion quickly escalated, with allegations being thrown around of those disagreeing about the predictions being "Intel fanboy", whose criticisms were demonstrative of having "their head in the sand". It finally ended with the user lamenting and concluding his opinions:
To which it was answered with
And now "next year" has come and...well...the facts speak for themselves.
Let this be a lesson about the futility in overhyping things. With Navi coming up, and then Zen 3 in mid-2020, let us be a bit more mature and not fall in love with every single preposterous rumor that pops up, however enticing they may be. Not all rumors are wrong, but we have now witnessed products, especially those by AMD (Polaris, Vega, Zen+, etc.), being overhyped time and time again, only for them to release with performance far below the initial illusory forecasts. Common sense is very clear on where we should stand and how we should behave.
Zen 2 is a great architecture and hardly any reasonable person would claim otherwise. Yet, compared to the predictions of the users above, Zen 2 is a disappointment. This is the sad irony in the erroneous confidence of these people: they end up damaging the very product they want to be good.
EDIT: I would like to underline that this discussion is about hypetrains and how we can avoid it. Let's please stick to the topic and not turn it into a trolling, baiting, flaming drama fest. Keep it civil, please.
AMD to Begin Sampling 7nm "Zen 2" Processors Within 2018 for a 2019 Launch
5775c had 128MB of eDRAM which gave it significant boost in some applications. No other CPU from Intel is using eDRAM, not even top of the line ones. I was hoping for Skylake to have eDRAM by default, but it didn't. Which sucked. I'm well aware of this processor architecture. Would love to see...
www.techpowerup.com
In May, 2018, several users made stupendous predictions about Zen 2's performance improvements, of which the predominant one was that of a user who claimed that a 20% IPC + 20% frequency + 50% core increase was well within range:
These opinions were of course disapproved by several users, as they displayed a severe lack of understanding of how technology works. Process node shrinks that do theoretically allow for these performance increases almost never really ever do so in practice (as virtually all cases of CPUs with node shrinks over the years have shown); the advertised improvements from node shrinks are always exaggarated, and never represent real-life cases. Nevertheless, the above user would have none of it, and also had his opinions vehemently defended by other participants:
The discussion quickly escalated, with allegations being thrown around of those disagreeing about the predictions being "Intel fanboy", whose criticisms were demonstrative of having "their head in the sand". It finally ended with the user lamenting and concluding his opinions:
To which it was answered with
And now "next year" has come and...well...the facts speak for themselves.
Let this be a lesson about the futility in overhyping things. With Navi coming up, and then Zen 3 in mid-2020, let us be a bit more mature and not fall in love with every single preposterous rumor that pops up, however enticing they may be. Not all rumors are wrong, but we have now witnessed products, especially those by AMD (Polaris, Vega, Zen+, etc.), being overhyped time and time again, only for them to release with performance far below the initial illusory forecasts. Common sense is very clear on where we should stand and how we should behave.
Zen 2 is a great architecture and hardly any reasonable person would claim otherwise. Yet, compared to the predictions of the users above, Zen 2 is a disappointment. This is the sad irony in the erroneous confidence of these people: they end up damaging the very product they want to be good.
EDIT: I would like to underline that this discussion is about hypetrains and how we can avoid it. Let's please stick to the topic and not turn it into a trolling, baiting, flaming drama fest. Keep it civil, please.
Last edited by a moderator:
