• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

Is the Pentium G4620 The First Decent Stopgap in Years?

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

crashtech

Diamond Member
Jan 4, 2013
9,638
1,520
126
Yes but with HT on, threads might be spread too far and between (so to speak) for a single core to have full load, especially if the GPU can't handle it's load(maxes out under the load before the/a cpu core can max out) .
If that is something that is occurring, it would certainly affect CPUs with high core counts disproportionately. Yet I'd think that even a DX12 game would still saturate at least one core with its main processes
 
Feb 25, 2011
16,562
1,326
126
One of my favorite tricks for building a nice rig on a budget back in the day was buy a new platform where everything was top notch- RAM, mobo, etc.- except for the CPU. The CPU would just be a stopgap until years later when you could get a top chip for that platform for cost savings that were greater than what the original stopgap CPU cost. The fact that almost every CPU could be OCed back then made waiting on the stopgap CPU not a painful experience.

Every since Sandy that practice seems foolhardy. Not only does it take longer for CPUs to fall off in value, but there hasn't really been a chip below a i3 that a serious user or gamer could survive on as a stopgap even with OC. You either went i5 or i7 day one or you were wasting money. That makes changing platforms very costly all at once, which pretty much forces some people to cut corners on say the mobo or RAM or GPU to afford that i5. We have needed a stopgap that wasn't there.

I tried to give the G3258 a chance, but it failed. It didn't matter how fast I could get the two cores, just having two threads meant even weak games like Rocket League stuttered. It was not the stopgap I was looking for.

But maybe the Pentium G4620 is finally a decent stopgap? Four threads means recent i3 performance in gaming, and even though its locked I bet it could still keep up with a 2500 non-k for the task.

Part of the question also comes down to cost. It looks like the Pentium G4620 will be sub-$100. Looking at 6700k prices, that chip was $370+ for months after launch, with some deal prices as low as $290 a year later when the demand had dropped. If history repeats that gives us a gap of $80 of possible savings if someone waits to buy a 7700k until its the best price possible. If the Pentium G4620 can be had for around $80, and its decent usage-wise, then there is no penalty for "getting by" on a Pentium G4620 for a year.

Is the Pentium G4620 the stopgap I dream it will be? Or will it fail gamers like the G3258, or have jacked up prices due to high demand for the early months of its existence? Thank you for any thoughts.
The concept of a "stopgap" CPU is, IMHO, flawed (as you suspect.)

Let's say the "useful lifespan" of an i7 CPU for gaming is six years. Simple question: if your options are "6 years of top notch gaming performance starting today" vs. "2 years of middling CPU performance, then a CPU swap, and then 4 years of top notch gaming performance." Is the first option worth $80 over 6 years?

To me, $80 is worth two years of not hobbling along with "keeping up with a 2500 non-K". I want an upgrade, dammit.

I'd rather skimp up front on the GPU or, hell, even storage, since those seem to depreciate more reliably and more rapidly, in terms of performance per dollar. Or recycle parts from a previous build until I can afford replacements.
 

Phynaz

Lifer
Mar 13, 2006
10,143
816
126
Or with HT,since basically it doubles cores as far as the cpu is concerned
No it doesn't. A CPU has the same number of cores regardless of SMT or not. a CPU doesn't see itself as having more cores.

Just because you can carry four grocery bags doesn't mean you have four hands as far as your body is concerned.
 

Shivansps

Diamond Member
Sep 11, 2013
3,140
797
136
That is what I figured but it was worth asking.
The only way this may work is, if Coffe Lake end up as 1151 and 200 mb compatible, it may be a good idea to get an Z270 and a G4560, then swap it for Coffe Lake 6C SKU.

This could be specially usefull for Nehalem/SB owners that are depending on very old MB, or even dying mbs.

If Coffe Lake compatibility is confirmed, i may sell my 2500K/Z67/16GB DDR3 and get Z270/G4560/32GB DDR4-3000... and later on drop a 6C Coffe Lake on it.
 

lopri

Elite Member
Jul 27, 2002
12,845
325
126
Even in the case of the quad HT should be called a tie or a win and not a regression; an increase in minimums can often indicate less stuttering, which is a real positive. The regression we see so often in hexacores is likely more due to OS scheduler limitations than anything to do with the CPU, of course the net result is still the same, I have HT off on my 5820K because of this.
That is an ironic reply considering that the benefits from HT comes in large part from the schedulers (both the CPU's and the OS'es) If HT were to claim a credit for performance gains which rely on the schedulers, then it should own up to the negatives caused by the schedulers as well.

Or with HT,since basically it doubles cores as far as the cpu is concerned,the cpu just doesn't see enough cpu usage to raise it's turbo clocks...
What HT does is filling underutilized data pipeline of a core by tricking the OS into thinking there are more resources (e.g. another core that in fact does not exists) than there actually are. Depending on the workload and the efficiency of application and the scheduler, there can be benefits and drawbacks. Think of it like a hair dresser who has an appointment every hour. She normally takes care of one customer every hour. But every few customers there comes one with a short hair who only needs trimming, and it takes only 15 minutes or so to do her/his hair. If that happens she will be done with that customer in 15 minutes and is free for 45 minutes, and if there are another customer or two who needs trimming she can do those customers' hair before her next appointment shows up.

As illustrated in this crude analogy, you can see that indeed there are scenarios in which the hair dresser can do more work than normally expected. But certain conditions have to be met. In her case: A customer with a lesser need for care than usual, a customer with a similarly lesser need, and the latter customer should be there when you are finished with the former so that she will not break her next appointment. If she miscalculates the time needed for each customers or some customers do not show up in time frame she is expecting, then other customers might have to wait past their appointment. (causing great anger and resentment in them)

HT fools the OS into believing there are two "logical" or "virtual" cores when there is one physical core. When that happens the OS will attempt to schedule two threads, and the second thread will be waiting in a state where it can be immediately worked on as soon as the first thread is done or stalled (e.g. needs to fetch new data). As you can imagine the potential gains or losses are highly dependent on a given situation as our hair dresser's gains/losses.

Yes but with HT on, threads might be spread too far and between (so to speak) for a single core to have full load, especially if the GPU can't handle it's load(maxes out under the load before the/a cpu core can max out) .
I think you have it backwards. Spreading the threads evenly to different physical cores, assuming they can be spread, is almost always the best practice for performance. Imagine a scenario where the OS schedule two threads into one physical core thinking there are two cores because of HT, while leaving an actual physical core idle. This actually happens a lot on desktop, especially in games. Bulldozer had a similar problem due to its integer units sharing a floating-point unit within a module: the OS would assign threads 1/2 on module 0, and threads 3/4 on module 1. On a 4 module / 8 thread CPU that meant the 4 threads had to fight for 2 modules' 2 floating-point units while leaving the other two modules idle. The performance hit was naturally enormous. An ideal practice for performance is assigning threads 1, 2, 3, 4 on modules 0, 1, 2, 3 separately so that each thread is given full resource of the integer unit and the floating-point unit in a module.
 
Last edited:

crashtech

Diamond Member
Jan 4, 2013
9,638
1,520
126
That is an ironic reply considering that the benefits from HT comes in large part from the schedulers (both the CPU's and the OS'es) If HT were to claim a credit for performance gains which rely on the schedulers, then it should own up to the negatives caused by the schedulers as well.
Okay, fine, but really what this digression into HT is about in regards to this thread is the fact that it works MUCH better on duals than quads and hexacores, something that many, including you, seem to have only become aware of recently. Perhaps your HT tutorial could come to some conclusions of why this is so in order to remain topical.
 

Ranulf

Golden Member
Jul 18, 2001
1,634
161
106
I'd rather skimp up front on the GPU or, hell, even storage, since those seem to depreciate more reliably and more rapidly, in terms of performance per dollar. Or recycle parts from a previous build until I can afford replacements.
Thats been my rational for several years. A video card is a lot easier to replace than a chip and is the one major part you will want to replace these days of 5 year cpus. Its not that hard to swap drives around (or just buy an ssd first and add on HDD later) either.
 

beginner99

Diamond Member
Jun 2, 2009
4,719
1,115
136
We don't need stop-gaps anymore. Just get the i7 or the i5 if you are cheap or very. very short on money and keep it for 5 years. So at $350 for the CPU that's $70 bucks per year.

The stop-gap solution was useful when CPUs actually made progress and we got new releases and uArch and not this borderline-rebrand shit. Skylake and kabylake suck to much for even ivybridge users to consider an upgrade or even 2600k owners. On the other hand those new CPUs also cost a leg and arm but price fell very quickly so it made sense to wait like it make sense to buy last years smartphone as you will save several $100. Totally worth it but sadly not for CPUs anymore.
 

beginner99

Diamond Member
Jun 2, 2009
4,719
1,115
136
This could be specially usefull for Nehalem/SB owners that are depending on very old MB, or even dying mbs.
I'm on lynnfield i7 still and I don't see how going for this Pentium would be a good solution. If my mobo fails I would just buy a 7700k (or Zen) and well wait it out again. Coffee-lake is at least 18 month away so i could easily upgrade then again if I want to.
 

Ancalagon44

Diamond Member
Feb 17, 2010
3,275
202
106
The concept of a "stopgap" CPU is, IMHO, flawed (as you suspect.)

Let's say the "useful lifespan" of an i7 CPU for gaming is six years. Simple question: if your options are "6 years of top notch gaming performance starting today" vs. "2 years of middling CPU performance, then a CPU swap, and then 4 years of top notch gaming performance." Is the first option worth $80 over 6 years?

To me, $80 is worth two years of not hobbling along with "keeping up with a 2500 non-K". I want an upgrade, dammit.

I'd rather skimp up front on the GPU or, hell, even storage, since those seem to depreciate more reliably and more rapidly, in terms of performance per dollar. Or recycle parts from a previous build until I can afford replacements.
This is unfortunately true, and the reason is that CPU performance has largely stagnated in the last couple of years. Just read all of the Kaby Lake reviews - there just isn't a massive change in performance in the last couple of years.

Compare that to GPUs. A modern low end GPU like the 1050 would actually compare favourably to a high end GPU from a few years ago, such as a GTX 580. But a Kaby Lake i3 is not going to wipe the floor with an Ivy Bridge i7. The i7 will still win, despite being years older and manufactured on a much older node.

You get more bang for your buck upgrading a GPU than upgrading a CPU.
 

StrangerGuy

Diamond Member
May 9, 2004
8,443
124
106
Let's say the "useful lifespan" of an i7 CPU for gaming is six years. Simple question: if your options are "6 years of top notch gaming performance starting today" vs. "2 years of middling CPU performance, then a CPU swap, and then 4 years of top notch gaming performance." Is the first option worth $80 over 6 years?

To me, $80 is worth two years of not hobbling along with "keeping up with a 2500 non-K". I want an upgrade, dammit.

I'd rather skimp up front on the GPU or, hell, even storage, since those seem to depreciate more reliably and more rapidly, in terms of performance per dollar. Or recycle parts from a previous build until I can afford replacements.
You are right. I never understood why people still thinks saving $100 by not going the i7 is a good idea ever since the 4790K for a part that lasts for 5 years. Heck, you can just recover that $80 you mentioned on the resale value of the i7 alone.

Now with the HT Pentiums, i5s makes less sense than ever.
 

Zstream

Diamond Member
Oct 24, 2005
3,397
277
136
I see three types of processors. The first being a low-end chip that's more than capable for browsing the web, YouTube, and Netflix. For this I use a AMD 5350, which is plenty fast.

The second is your mid-market tier - bang for the buck. Like a haswell v3 1231.

The third is your top tier - best CPU on the market.

I honestly don't see why people go and find in between processors. Pick a tier and be happy.
 

VirtualLarry

No Lifer
Aug 25, 2001
50,419
6,025
126
Well, I'm in for a G4600.

Edit: Was originally going to go for the G4560, largely because it's cheaper, but my limited experience was that the i3-6100 was faster than the G4400 for web browsing, and supposition that part of that was the faster iGPU. So with the G4600 having the HD630, and being available NOW for $89, it won out over waiting for a hypothetical $64 G4560 on the 16th.
 
Last edited:
  • Like
Reactions: crashtech

daxzy

Senior member
Dec 22, 2013
393
77
101
Well, I'm in for a G4600.

Edit: Was originally going to go for the G4560, largely because it's cheaper, but my limited experience was that the i3-6100 was faster than the G4400 for web browsing, and supposition that part of that was the faster iGPU. So with the G4600 having the HD630, and being available NOW for $89, it won out over waiting for a hypothetical $64 G4560 on the 16th.
Whats the difference between the HD610 and HD630?
 
Feb 25, 2011
16,562
1,326
126
Well, I'm in for a G4600.

Edit: Was originally going to go for the G4560, largely because it's cheaper, but my limited experience was that the i3-6100 was faster than the G4400 for web browsing, and supposition that part of that was the faster iGPU. So with the G4600 having the HD630, and being available NOW for $89, it won out over waiting for a hypothetical $64 G4560 on the 16th.
Are these going into your G4400 rigs as "upgrades", or are you building two more boxes?
 

VirtualLarry

No Lifer
Aug 25, 2001
50,419
6,025
126
LOL. Yes and no. I'm only getting one CPU, thus far, and there are three (well, more) boxes it could go in.

I was primarily interested in it going into one of my DeskMini PCs, where it should pretty-much stay.

But I do have an interest in attempting to OC it in my ASRock Z170 Pro4S board(s).

And I do have a fresh new mobo to put a 1151 CPU into, which I was thinking of my friend, he's been wanting an upgrade for a while now (but has no money, because he spends it on non-computer things). Then wonder why he has such a (poor) PC.
 

ASK THE COMMUNITY