Flash forward to better times? Intel is said to have increased Arrow Lake clocks to 5.3-5.5 GHz

High expectations

Processor generation Arrow Lake was supposed to be essentially a revolution. First major architecture change since release Alder Lake in 2021 and actually with an even more significant shift. The first deployment of tiles in the desktop (at least since it was rumored that Meteor Lake will only be a mobile solution). The first deployment of the Intel 20A process, so a huge leap compared to the previous Intel 7 process. Intel’s first standard desktop processor with more than 24 cores (8+32 cores configuration).

Unfavorable development of the situation

However, great expectations were dashed only for a large increase in IPC, as rumors began that HT (two threads per core) broke down as a result of architectural changes, while the considered alternative in the form of so-called profitable units (combining the performance of two cores into one thread) was put into operation failed. Most sources have also reported that Intel is struggling with clock speeds that can’t even come close to the last generation (reminiscent of the situation with generations Cannon Lake / Ice Lake). This was followed by reports that the configuration with an increased number of cores (8+32) has problems and will either be canceled or delayed, in any case, the number of cores will remain at the current 24 (8+16). It was also confirmed that neither Arrow Lake it won’t support AVX-512, as small cores can’t handle it.

Favorable news

After a long time, more positive information finally appeared. According to long-time leaker MebiuW, Intel managed to increase the maximum clock frequency Arrow Lake at 5.3-5.5 GHz. Although this is less than the 6.2 GHz u Raptor Lake-refresh, but it is no longer a problem that clocks at the level of 4.x GHz would be. It is unclear whether Intel managed to achieve these results by switching to TSMC’s 3nm process from the original Intel 20A or by tweaking the chip design.

Source: DIIT

Now it’s all about IPC. According to most sources, it should increase by 30%+ intergenerationally, according to some it is more like ~15%. If Intel really managed to jump by at least 30%, then 5.3-5.5 GHz clocks will not be a problem, because the performance (including the drop in clocks) will shift between generations by at least ~20%, hypothetically similar to AMD, which when consumption drops (it is expected that PL2 limit Arrow Lake will be 177 watts thanks to TSMC’s 3nm process) can look very good. If the IPC were to move more like 15%, then the intergenerational increase in performance would not be very significant (percentage units), which combined with the need to purchase a new platform would Arrow Lake versus Zen 5 rather uninteresting.

But the reality can be somewhere in between. For a long time, information is heard that small cores Arrow Lake they will bring a higher offset than large ones, so a 30% increase in IPC can be for small cores or overall performance (where the effect of small cores predominates) and a 15% increase in single core performance (ie large cores).

Source: DIIT

Another piece of news came from Golden Pig Upgrade, a hardware vendor and leaker who received information that Thunderbolt 4 will become standard on motherboards with the high-end Z890 chipset, and that integrated graphics Arrow Lake for desktop it will have 4 active Xe blocks, but variants with 2-3 active Xe blocks may also appear.

Finally, it started to be rumored that Intel is doing everything in order to release Arrow Lake happened as soon as possible. That is, not until the very end of the year, as was the case with Meteor Lake. It was previously mentioned that he could talk about this generation already at Computex (end of May / June), but now there are mentions that he would like to come up with something before the end of the third quarter, i.e. before the end of September. Whether it would be a formal launch or the availability of some models, but it is not clear.