Yes, it conveys a clear idea about balance—strong foundation with minimal central focus.
Yes, it conveys a clear idea about balance—strong foundation with minimal central focus.
This clearly makes sense from a gaming standpoint. You rely on strong processors for performance, while background tasks run smoothly on less powerful cores. If you're working on video editing and your CPU is split between efficient cores, it can slow down rendering. Those high-performance cores need to be exceptional to compensate, or they'll just be unused capacity. Personally, I’d pick the Ryzen 9 5950X with its many powerful cores over the Intel i9 12890KFC, even if the latter is cheaper.
It functions when battery life and heat management matter. Results on mobile/AIO/laptops depend on one or both of these aspects. This approach isn't as critical on desktop systems where power and cooling can be optimized, which is why it's absent from desktop designs.
I believe mainstream platforms should remain distinct from HEDT systems. big.LITTLE fits well for everyday use, similar to how cylinder deactivation works in vehicles—providing power when needed and saving energy otherwise. This approach is especially useful for laptops and smartphones, where the latter has already demonstrated the value. For desktops, it offers a way to boost chip performance and efficiency without a large cost increase, especially since the platform is already tailored for mobile devices. Just as you wouldn't apply cylinder deactivation to a commercial generator, it shouldn't be used in HEDT systems where the advantages wouldn't be realized.
Well, yes... Intel seems to be implementing similar changes in their HEDT products too. Cylinder deactivation was a problem before, and companies like Chrysler, Ford, and GM tried it back in the 80s and 90s. It worked in theory but turned out very complicated, didn’t really deliver the promised efficiency, and was unreliable. For CPU cores it’s different, but the trend is concerning. If Intel adopts this approach with their HEDT systems, they might permanently lose competitive edge to AMD. That would hurt competition, which is something I understand is important. I’m not too worried about Intel struggling, just that it could change the market dynamics.
Some discussions on forums clarified that power limits affect desktop CPUs too. For instance, the 5950X uses the most power when running at 8-10 cores even though it has 16 cores. Running with fewer cores could free up more power for demanding multithreaded tasks.
This choice wouldn’t ignore the limitation of having fewer total threads, nor would it overlook the fact that half your threads aren’t top-tier performance. I’d quickly opt for a BC/LC mobile in such a case, but not for my video editing setup. My rig features a Ryzen 9 and an older Xeon—12 cores versus 4 and 8 respectively. The Ryzen runs efficiently at idle, drawing only about 100 watts from the wall, while the Xeon plus its monitor consumes nearly 160 watts. Although the Xeon might outperform it under heavy load, I’d rely on the Xeon to sustain high performance for extended periods, as demonstrated by my 66-hour transcoding session. I wouldn’t replace the Xeon with a BC/LC device, but I see value in keeping it for its reliability. The real concern is whether Intel would adopt BC/LC across all systems, including high-end models—if they do, it could create significant challenges. My priority isn’t idle power savings; it’s maintaining consistent performance when it matters most.
Many vehicles and trucks still rely on it without realizing it. These components have been resolved. Regarding Intel's implementation in HEDT... We should hold off until an official statement comes out. It could be useful in power-constrained situations, but we've seen how the Xeon Phi performed.
The issue raised concerns about large cores struggling to perform optimally when multiple cores are active. This raises questions about their necessity if not all can be utilized simultaneously. Moreover, smaller cores typically feature a reduced die size, suggesting a shift from having numerous large cores to combining larger and smaller ones on the same chip.
The visuals from Intel appeared to display four main cores followed by six or eight smaller ones. This could be misleading. A Reddit thread lists vehicles with DOD, but I wouldn’t consider it a major trend. DOD isn’t performing well even on four-cylinder engines, though Ford does use a three-cylinder model with it. It seems more accurate to say that trucks and big SUVs often use DOD. Another issue is that DOD might improve economy slightly, but unless you’re running a large V8, it doesn’t significantly boost overall performance. Using a smaller engine with a turbo and proper tuning can deliver better efficiency and power when needed. To my view, Intel seems to limit the chip’s top performance potential, whereas I’d favor a design that maximizes peak performance while still offering good economy. Personally, the amount of power isn’t crucial unless you’re editing videos; otherwise, modern chips already provide excellent idle performance, making halving idle draw negligible for me.