The main requirement in an enterprise network is that all clocks are synchronised as that will allow you to correctly correlate (security) events found in different systems. Additionally some protocols rely on time-stamps and will fail when there is too much time offset between systems.
There is an old saying that goes: "A man with one watch knows what time it is. A man with two watches is never sure."
If you have only one clock, that is the only one you can trust. If you have two and they start to show different times, it is difficult to know which one has gone wrong. To guarantee accuracy, you need at least three clocks. If two of them show the same time, you can be relatively sure when the third clock has gone wrong. The more clocks you have showing the same time, the more sure you can be that they are right and that any clock showing a different time is incorrect.
source: RIPE: Christer
Weinigel
Generally that means that when you environment already has NTP servers, you should have at least three and that you configure all systems to use all of those NTP servers as their time source.
When you only provide a single NTP source to clients, regardless if that is only a single one of your central NTP servers or the gateway IP-address, you deprive your clients of the ability confirm the accuracy of that clock and their clock can much more easily become inaccurate in case of failure in the gateway clock.
Arguably you could consider a tiered approach, your network devices and routers synchronise their clocks with the NTP servers and your devices synchronise their clocks only with their gateway routers. That might be useful with large numbers of badly behaved devices that would otherwise overload your central NTP servers. In almost all other cases I think that such a tiered approach should be avoided.
